Managing Oracle Database Systems User Guide

Transcript

Foglight™ for Oracle Managing Oracle Database Systems User and Reference Guide © 2014 Quest Software, Inc. ALL RIGHTS RESERVED. This guide contains proprietary information protected by copyright. The software described in this guide is furnished under a software license or nondisclosure agreement. This software may be used or copied only in accordance with the terms of the applicable agreement. No part of this guide may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording for any purpose other than the purchaser’s personal use without the written permission of Quest Software, Inc. The information in this document is provided in connection with Quest products. No license, express or implied, by estoppel or otherwise, to any intellectual property right is granted by this document or in connection with the sale of Quest products. EXCEPT AS SET FORTH IN QUEST'S TERMS AND CONDITIONS AS SPECIFIED IN THE LICENSE AGREEMENT FOR THIS PRODUCT, QUEST ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY EXPRESS, IMPLIED OR STATUTORY WARRANTY RELATING TO ITS PRODUCTS INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. IN NO EVENT SHALL QUEST BE LIABLE FOR ANY DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDENTAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF PROFITS, BUSINESS INTERRUPTION OR LOSS OF INFORMATION) ARISING OUT OF THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF QUEST HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Quest makes no representations or warranties with respect to the accuracy or completeness of the contents of this document and reserves the right to make changes to specifications and product descriptions at any time without notice. Quest does not make any commitment to update the information contained in this document. If you have any questions regarding your potential use of this material, contact: Quest Software World Headquarters LEGAL Dept 5 Polaris Way Aliso Viejo, CA 92656 www.quest.com email: [email protected] Refer to our Web site for regional and international office information. Trademarks Quest, Quest Software, the Quest Software logo, Foglight, IntelliProfile, PerformaSure, Spotlight, StealthCollect, TOAD, Tag and Follow, Vintela Single Sign-on for Java, vOPS, and vFoglight are trademarks and registered trademarks of Quest Software, Inc in the United States of America and other countries. For a complete list of Quest Software’s trademarks, please see http://www.quest.com/legal/trademark-information.aspx. Other trademarks and registered trademarks are property of their respective owners. Third Party Contributions Foglight contains some third party components. For a complete list, see the License Credits page in Foglight online help. User and Reference Guide June 2014 Management Server version 5.6.11 Cartridge version 5.6.10.3 Managing Oracle Database Systems: User and Reference Guide 3 Table of Contents Introduction to this Guide..................................................................................................................................................7 About Quest Software, Inc. ............................................................................................................................................................. 7 Contacting Quest Software ..................................................................................................................................................... 7 Contacting Quest Support ....................................................................................................................................................... 8 Introducing Foglight for Oracle.........................................................................................................................................9 Using Foglight for Oracle.................................................................................................................................................10 Upgrading to the Current Version.................................................................................................................................................. 10 Overview .............................................................................................................................................................................. 10 Running the Upgrade Wizard ................................................................................................................................................ 12 Adding and Configuring Agents..................................................................................................................................................... 14 Configuring Multiple Instances for Monitoring using CLI ....................................................................................................... 15 Reviewing the Foglight for Oracle Architecture .................................................................................................................... 21 Configuring RMI Server on the Agent Manager Concentrator .............................................................................................. 22 Running the Oracle Monitoring Wizard ................................................................................................................................. 24 Adding Agents for Monitoring Exadata Environments........................................................................................................... 42 Introducing the Foglight for Oracle Drilldowns .............................................................................................................................. 44 Overview ............................................................................................................................................................................... 44 The Foglight for Oracle Layout...................................................................................................................................................... 45 Reviewing the Databases Dashboard .................................................................................................................................. 47 Selecting the Instance to be Monitored ................................................................................................................................. 54 Foglight for Oracle Overview Page ....................................................................................................................................... 57 Reviewing Performance-related Data ................................................................................................................................... 59 Reviewing Overall Activity ..................................................................................................................................................... 96 Reviewing RAC Historic Activity.......................................................................................................................................... 103 Reviewing Instance Historic Activity.................................................................................................................................... 109 Viewing the Foglight for Oracle Real-time Summary Page ................................................................................................. 113 Oracle Activity Drilldown ..................................................................................................................................................... 122 Pluggable Databases Drilldown .......................................................................................................................................... 135 Storage Drilldown ............................................................................................................................................................... 136 Reviewing Configuration Settings ....................................................................................................................................... 151 Reviewing the Alert and Error Logs..................................................................................................................................... 155 Reviewing Exadata-related Information .............................................................................................................................. 157 Foglight for Oracle Agents........................................................................................................................................................... 161 Managing Oracle Database Systems: User and Reference Guide 4 Table of Contents Working with Agent Properties............................................................................................................................................. 161 DB_Oracle Agent Overview ................................................................................................................................................ 162 DB_Oracle Agent Properties ............................................................................................................................................... 162 DB_Oracle_RAC Agent Overview ...................................................................................................................................... 162 DB_Oracle_RAC Agent Properties ..................................................................................................................................... 163 DB_Oracle_RAC_Instance Agent Overview ....................................................................................................................... 163 DB_Oracle_RAC_Instance Agent Properties ..................................................................................................................... 163 DB_Oracle_ASM Agent Overview ...................................................................................................................................... 164 DB_Oracle_CRS Agent Overview ...................................................................................................................................... 164 DB_Oracle_Exadata Agent Overview.................................................................................................................................. 164 Managing Foglight for Oracle Agent Settings .............................................................................................................................. 164 Opening the Databases Administration Dashboard............................................................................................................. 165 Reviewing the Administration Settings................................................................................................................................. 165 Defining Connection Details................................................................................................................................................. 166 Defining Data Collection and Storage Options .................................................................................................................... 173 Customizing Alarms for Foglight for Oracle Rules ............................................................................................................... 175 Defining Retention Policies .................................................................................................................................................. 187 Defining the Collection Settings ........................................................................................................................................... 188 Configuring the Alert Log Panel Display .............................................................................................................................. 190 Configuring the List of Listeners per Instance...................................................................................................................... 192 Configuring the Connection to Performance Analysis ......................................................................................................... 193 Configuring User-defined Collections .................................................................................................................................. 194 Configuring the On-demand Data Port ................................................................................................................................ 197 Changing the Monitoring Mode of the RAC One Node Agents............................................................................................ 198 Generating Reports ..................................................................................................................................................................... 199 Generating Reports for a Foglight for Oracle RAC or Instance ........................................................................................... 199 Studying the Various Reports .............................................................................................................................................. 199 Glossary....................................................................................................................................................................................... 200 Performance Analysis Metrics ..................................................................................................................................................... 207 Reference........................................................................................................................................................................ 222 Foglight for Oracle Rules ............................................................................................................................................................. 222 Overview of Collections .............................................................................................................................................................. 224 Advisories ............................................................................................................................................................................ 224 Alert Log............................................................................................................................................................................... 225 Archive Destination Miscellaneous ...................................................................................................................................... 226 Archive Logs Storage........................................................................................................................................................... 227 ASM Diskgroups .................................................................................................................................................................. 229 ASM Diskgroups Balance .................................................................................................................................................... 231 ASM Disks ........................................................................................................................................................................... 231 ASM Instance Total.............................................................................................................................................................. 233 Managing Oracle Database Systems: User and Reference Guide 5 Table of Contents ASM Operations................................................................................................................................................................... 234 Backup Config...................................................................................................................................................................... 235 Backup Jobs ........................................................................................................................................................................ 236 Backup Type Status............................................................................................................................................................. 237 Change Tracking.................................................................................................................................................................. 237 Cluster Latency.................................................................................................................................................................... 238 Cluster Misses ..................................................................................................................................................................... 242 Datafile Fragmentation......................................................................................................................................................... 243 Datafile IO Activity................................................................................................................................................................ 243 Datafiles Storage ................................................................................................................................................................. 245 Dataguard Destination Status .............................................................................................................................................. 247 Dataguard Redo Apply......................................................................................................................................................... 248 Dataguard Status................................................................................................................................................................. 249 Dump File............................................................................................................................................................................. 250 Event Wait............................................................................................................................................................................ 251 Exadata Cell Flashcache ..................................................................................................................................................... 251 Exadata Cell General........................................................................................................................................................... 253 Exadata Cell InfiniBand........................................................................................................................................................ 254 Exadata Cell InfiniBand Ports .............................................................................................................................................. 255 Exadata Cell InfiniBand Switches ........................................................................................................................................ 256 Exadata Cell Statistics ......................................................................................................................................................... 257 Exadata Cell General........................................................................................................................................................... 258 Filesystem Disk.................................................................................................................................................................... 259 Flashback Area.................................................................................................................................................................... 260 History Locks Tree Instance ................................................................................................................................................ 261 History Locks Tree RAC ...................................................................................................................................................... 263 Host Properties .................................................................................................................................................................... 265 Initialization Parameters....................................................................................................................................................... 266 Instance General.................................................................................................................................................................. 268 Invalid Objects ..................................................................................................................................................................... 268 Invalid Objects List............................................................................................................................................................... 270 IO Activity............................................................................................................................................................................. 270 Jobs Queue.......................................................................................................................................................................... 277 Latches Activity.................................................................................................................................................................... 278 Locks Activity ....................................................................................................................................................................... 279 Locks Breakdown................................................................................................................................................................. 280 Locks Instance..................................................................................................................................................................... 281 Locks Tree ........................................................................................................................................................................... 282 Locks Tree RAC................................................................................................................................................................... 284 OS and Instance Statistics................................................................................................................................................... 285 OS File System IO Summary............................................................................................................................................... 292 Managing Oracle Database Systems: User and Reference Guide 6 Table of Contents OS General Network Summary ........................................................................................................................................... 293 PA Instance Statistics .......................................................................................................................................................... 293 PA Usability.......................................................................................................................................................................... 296 Parallel ................................................................................................................................................................................. 297 PGA Statistics...................................................................................................................................................................... 298 Pluggable Databases Information........................................................................................................................................ 300 Pluggable Databases State ................................................................................................................................................. 301 Redo Archive Performance.................................................................................................................................................. 301 Redo Instances Miscellaneous ............................................................................................................................................ 303 Redo Logs Member Storage................................................................................................................................................ 304 Redo Log Storage................................................................................................................................................................ 304 Session Details Datafile I/O PA ........................................................................................................................................... 305 Session Details Locks PA .................................................................................................................................................... 306 Session Details Open Cursors PA ....................................................................................................................................... 307 Session Details SQL Summary PA...................................................................................................................................... 308 Session Details Statistics PA ............................................................................................................................................... 309 Session List PA.................................................................................................................................................................... 312 Sessions Summary .............................................................................................................................................................. 323 Servers................................................................................................................................................................................. 325 Subcategories Wait.............................................................................................................................................................. 325 Tablespaces Free Space Fragmentation............................................................................................................................. 330 Tablespaces Storage........................................................................................................................................................... 331 Top Blocked Objects............................................................................................................................................................ 334 Top Sessions ....................................................................................................................................................................... 335 Top SQLs............................................................................................................................................................................. 338 Top SQLs Full Text.............................................................................................................................................................. 340 Top SQLs Text..................................................................................................................................................................... 340 Total Archive Storage .......................................................................................................................................................... 341 Total Database Storage....................................................................................................................................................... 341 Total Instance Waits ............................................................................................................................................................ 342 Undo Activity ........................................................................................................................................................................ 343 Total Summary..................................................................................................................................................................... 344 Usability ............................................................................................................................................................................... 345 Glossary....................................................................................................................................................................................... 347 Index................................................................................................................................................................................ 350 Introduction to this Guide The Foglight for Oracle User and Reference Guide provides configuration instructions, conceptual information and instructions on how to use Foglight for Oracle and its component agents. This guide is intended for Oracle Database Administrators, who want to use Foglight for Oracle for monitoring their environments. This guide covers the entire monitoring process, starting with the addition of the database instances and the connection with these instances, and proceeding with the use of the various drilldowns and the Databases Administration dashboard options. About Quest Software, Inc. Established in 1987, Quest Software (Nasdaq: QSFT) provides simple and innovative IT management solutions that enable more than 100,000 global customers to save time and money across physical and virtual environments. Quest products solve complex IT challenges ranging from database management, data protection, identity and access management, monitoring, user workspace management to Windows management. For more information, visit www.quest.com. Contacting Quest Software Email [email protected] Mail Quest Software, Inc. World Headquarters 5 Polaris Way Aliso Viejo, CA 92656 USA Web site www.quest.com Refer to our Web site for regional and international office information. Managing Oracle Database Systems: User and Reference Guide 8 Introduction to this Guide Contacting Quest Support Quest Support is available to customers who have a trial version of a Quest product or who have purchased a Quest product and have a valid maintenance contract. Quest Support provides unlimited 24x7 access to our Support Portal at https://support.software.dell.com. From our Support Portal, you can do the following: • Retrieve thousands of solutions from our Knowledge Base • Download the latest releases and service packs • Create, update, and review Support cases View the Global Support Guide for a detailed explanation of support programs, online services, contact information, policies, and procedures. The guide is available at: https://support.software.dell.com/images/Global-Support-Guide.pdf. 1 Introducing Foglight for Oracle The Foglight for Oracle cartridge enables Foglight for Oracle to monitor the Oracle database activity by connecting to, and querying, the Oracle instances within an enterprise. Oracle database data is monitored through agents. For more information, see “Foglight for Oracle Agents” on page 161. Starting to work with Foglight for Oracle requires first adding Oracle database instances and connecting to these instances. These operations are carried out using the Oracle Monitoring Installer wizard. For more information, see “Adding and Configuring Agents” on page 14. If Foglight for Oracle has been upgraded to the latest version, starting to work with a Foglight for Oracle instance requires upgrading to the current version both the agent and the Foglight Agent Managers that host the agents. For more information, see “Upgrading to the Current Version” on page 10. The status and performance of the Oracle instances can be viewed on the Oracle dashboards. For more information, see “Introducing the Foglight for Oracle Drilldowns” on page 44. Setting options for collecting, storing, and displaying data is carried out using the Databases Administration dashboard. For more information, see “Managing Foglight for Oracle Agent Settings” on page 164. Use the Reports dashboard to generate reports about various performance-related aspects of the selected instance or RAC. For information on how to generate the various reports, as well as a brief description of each report, see “Generating Reports” on page 199. 2 Using Foglight for Oracle Foglight for Oracle monitors the Oracle database activity by connecting to and querying the Oracle database. The agents provided monitor the Oracle database system. The dashboards included with the cartridge provide a visual representation of the status of the major components of the Oracle RAC and instance agents. They allow you to determine any potential bottleneck in database performance. Upgrading to the Current Version Starting to work with a Foglight for Oracle instance requires upgrading to the current version both the instance and the Foglight Agent Manager that runs the instance. This chapter contains instructions for using the upgrade wizard. Note This chapter is relevant only for existing Foglight users; if Foglight for Oracle is being used for the first time, and the version of Agent Manager being used is the updated version, as indicated in the release notes, skip this chapter. Important Foglight for Oracle does not support the upgrade method of placing the new version under the folder FGLHOME/upgrade/cartridge. Important If upgrading to the current version of Foglight for Oracle in a Federation architecture, the upgrade should be applied first to the federated database servers and then to the Federation Server. If the Federation Server is upgraded first, this server will display incorrect information regarding the number of instances being monitored, as reported under the Status Summary section. All the same, this information will be displayed correctly on the federated databases. Overview Any upgrade of Foglight for Oracle to the latest version, which is not met by similar upgrade of the database cartridge components, is detected after the Databases dashboard's display is being loaded, either automatically (by default: every 60 seconds) or manually, by clicking F5. Upon detecting the need for upgrade, the upgrade wizard starts automatically. Managing Oracle Database Systems: User and Reference Guide 11 Chapter 2—Using Foglight for Oracle Selecting the check box at the bottom left of the screen prevents this wizard from appearing when entering the Databases dashboard. Nevertheless, if several database cartridge components still require upgrade, the need to upgrade them is indicated in the dashboard. The indication is evident both in the caption Upgrade required, which is displayed in red to the right of the requested instance or RAC, and in the button Upgrade, which appears only if upgrade is required. Note Even though instances with components that need to be upgraded appear in the Databases table, such instances cannot be accessed by clicking them, until they have been fully upgraded. Upgrade button Managing Oracle Database Systems: User and Reference Guide 12 Chapter 2—Using Foglight for Oracle Running the Upgrade Wizard To upgrade the requested instances: 1 Click anywhere on a row that displays the Upgrade required status. Alternatively, click the Upgrade button. The screen that appears now is the same screen that appears by default upon entering the Databases dashboard when one or more database cartridge components require upgrading. While the need for upgrade is indicated for both Foglight for SQL Server and Foglight for Oracle, the upgrade is carried out separately for each database cartridge type. 2 Click Foglight for Oracle. 3 Select the Agent Managers to upgrade. Note The Other Agents column in the table indicates whether the specified Agent Manager runs agents of other cartridge types. If this column holds a value, navigate to Dashboards > Administration > Agents > Agent Status to view which other agents run under the specified Agent Manager. If the other agents appear in the list below, upgrading Agent Manager to the latest version is safe. For agents of any other cartridge type, consult the documentation of the respective cartridge types. DB_DB2_* DB_SQL_Server_* DB_Oracle_* Sybase_MDA UnixAgent WindowsAgent 4 Click Upgrade the selected agent managers. A progress bar appears. Managing Oracle Database Systems: User and Reference Guide 13 Chapter 2—Using Foglight for Oracle 5 After the upgrade is complete, click Next. The next screen is used for granting privileges to users that were detected as possibly requiring additional privileges to ensure full functionality. 6 Select the agents displayed in the table and click Validate connectivity. 7 If the status row of one or more agents displays the status Insufficient privileges, complete the following steps: a Select all agents that require the requested privileges, and click Grant privileges. b In the Grant Database Privileges dialog box, enter a SYSDBA user and password. c To view the script used for granting the privileges, click View grant script. Important The script for granting privileges is invoked using a popup. To view the script, ensure that pop ups are not blocked on the page. d Click Grant. Managing Oracle Database Systems: User and Reference Guide 14 Chapter 2—Using Foglight for Oracle 8 On the Rule Modification Overview screen, review the list of rules added, removed, or updated. Rules have one of the following states: • Removed — The rule is obsolete and was deleted from the Management Server. • New — The rule is new in this release. To review its definition, click View new rule. • Updated — The rule was updated in this release. To review the updated rule, click View updated rule. If a rule targeted for update was modified by a user, the user’s modified rule is copied and disabled before the updated rule is installed. To review the modified rule, click View user-updated rule. Tip To avoid having to repeat rule modifications after an upgrade, do not edit rules with the DBO prefix in the Rule Management dashboard. Use the Alarms screen on the Databases Administration dashboard instead. When you make your edits on the Alarms screen, the edits are saved separately and applied over the predefined rules. For instructions, see “Customizing Alarms for Foglight for Oracle Rules” on page 175. Adding and Configuring Agents Enabling Foglight Management Server to monitor Oracle instances requires creating the Foglight agents that monitor these instances and ensuring that these agents communicate properly with the Management Server. Foglight for Oracle provides Oracle monitoring wizard - a graphical, intuitive method for creating and configuring multiple agents, which can be used instead of the default method offered by Foglight for creating agents and editing their properties using the Agent Administration view (see chapter Foglight for Oracle Agent in the Foglight for Oracle Reference Guide). This chapter contains instructions for using the Oracle monitoring wizard, as well as the steps that should be taken for monitoring the newly added database instances. For details, see “Running the Oracle Monitoring Wizard” on page 24. Important When running Foglight for Oracle in a Federation architecture, neither the creation nor the administration of agents can be accomplished from the central Management Server (the Managing Oracle Database Systems: User and Reference Guide 15 Chapter 2—Using Foglight for Oracle Federation Master). These two tasks should be carried out from the stand-alone Management Servers (the Federated Children). Note In addition to using a Wizard, it is possible to configure multiple instances for monitoring through a silent installation that is carried out using the command line. For details, see “Configuring Multiple Instances for Monitoring using CLI” on page 15. Configuring Multiple Instances for Monitoring using CLI Note The instructions below refer to adding multiple Oracle agents. A similar process exists for adding multiple ASM instances. For more information, see “Configuring Multiple Instances for ASM” on page 19. To add multiple agents using the silent installation through a command line interface: 1 In the navigation panel, under Dashboards, click Administration > Cartridges > Components for Download. 2 Click the Installer icon beside DB_Oracle_CLI_Installer. 3 Follow your browser’s prompts to save the ZIP file. 4 Extract the ZIP file to a directory of your choice. The ZIP file contains the following files: • README.txt file • oracle_cli_installer.groovy — a groovy script file that runs the silent installation. This file should be copied to the /bin directory. • oracle_silent_installer_input_template.csv — a template file that should serve as the basis for inserting contents into the input CSV file. This file can be copied to any folder of your choice, provided that the path indicated by the parameter (see below) points to the selected path. For details about the contents of this file, see “Contents of the Input CSV File” on page 16 5 Copy the oracle_cli_installer.groovy file to the /bin directory. 6 Go to the command line and execute the command: /bin/fglcmd -srv -port -usr -pwd -cmd script:run -f oracle_cli_installer.groovy fglam_name oracle_instances_file_name lockbox_name lockbox_password The descriptions of the flags and parameters in this command are as follows: • • • • • : The Foglight Management Server installation directory. : The name of the host where the Management Server is installed. : The Management Server HTTP port. : The user name used for connecting to the Management Server. : The password of the specified user. Managing Oracle Database Systems: User and Reference Guide 16 Chapter 2—Using Foglight for Oracle • : The full path and the name of the input CSV file. • : Optional — specifies the name of an existing lockbox that would be used. • : Optional — the selected lockbox's password. 7 After the installation process completes, review the CSV files that are created in the same folder where the input file resides: • A file with the input file's name and _status suffix (for example: if the input file is named input, this file is named input_status). This file indicates the monitoring situation of all Oracle instances listed in the input CSV file. If monitoring failed for one or more of the instances, the file specifies the error message. The _status file includes the name of the monitored Foglight for Oracle agent, the result of the monitoring validation process - MONITORED, FAILED, or AGENT EXISTS - and the error message, in case the agent creation failed. Important If all instances listed in the input CSV file passed the monitoring validation successfully, no additional file is created. • A file with input file's name and _new suffix (for example: if the input file is named input, this file is named input_new). This file, which is only created if one or more instances, nodes or RACs failed the monitoring validation, contains all details about these instances, RACs, or nodes. Using this CSV file, the user only needs to fix the errors and re-run the script, this time by specifying this file as the input file, that is: the original file name with suffix _new. In the example used here, the new file should be named input_new.csv. Contents of the Input CSV File The input CSV file contains the following fields, which are used as columns in the resulting file: Name Description RAC Agent Name The name of the RAC agent. This name must be unique per agent. • If the instance belongs to a RAC, this field has to contain the name of the RAC agent. • If this RAC runs multiple nodes, all instances must have the same RAC agent name. • If the instance does not belong to a RAC, leave this field empty. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Name Description Instance Type The type of the instance to be monitored. The possible values are: SINGLE, NODE, RAC ONE NODE. • SINGLE: the instance does not belong to a RAC. • NODE: the instance belongs to a RAC that has more than one instance. • RAC ONE NODE: the instance belongs to a RAC with a single node. Agent Name The requested agent name. This name must be unique per agent. If an agent with the given name already exists, the newly provided agent will not be created. Host The host that is serving the Oracle instance or RAC. Port The TNS listener port. Connection Method The requested connection method. The possible values are either SERVICE_NAME or SID. SID/Service Name The name of the SID or service name used for establishing connection to the instance. User name The name of the user to be used for connecting to the instance. Note The user name for monitoring must exist prior to running the silent installation. Password The password of the database user name. Administrator User Name (Optional) The user name of a user that can grant privileges to the regular database user; required only if the regular user does not have sufficient privileges. Administrator Password (Optional) The password of the database administrator user name mentioned above. Enable OS Monitoring Specifies whether to create an Infrastructure cartridge agent for the given host name; the possible values are True or False. OS Platform (Optional) The type of the OS platform where the Oracle instance runs; the possible values are either UNIX or WINDOWS. 17 Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Name Description Authentication (Optional) OS authentication type. This field can have one of the following values: • LOGIN_PASSWORD: user/password authentication • RSA_KEY: RSA-based public key authentication • DSA_KEY: DSA-based public key authentication • CLIENT_LOGIN: uses the Foglight Agent Manager's login • WINDOWS_ACCOUNT: Windows authentication, in the domain\user and password convention OS User Name (Optional) The user name required for connecting to the OS. OS Password (Optional) The password of the OS user specified above. SSH Port (Optional) The port that is used for listening to SSH connections on the monitored hosts; if the monitored host is Windows-based, leave this field empty. Private Key File Path (Optional) The full path of the private key file; relevant only to the following credential types: RSA_KEY, DSA_KEY. Passphrase (Optional) The passphrase of the OS user that is used in public key authentication; relevant only to the following credential types: RSA_KEY, DSA_KEY. Use sudo (Optional) Used for indicating that the OS user specified above is not an administrator user, but can run certain commands that require administrative privileges as root. The possible values are TRUE or FALSE. if the monitored host is Windows-based, leave this field empty to indicate FALSE. Enable VMware Collection (Optional) Indicates whether to monitor VMware metrics; The possible values are TRUE or FALSE. VMware Host (Optional) The VMware host name. VMware Port (Optional) The VMware port. If no value is inserted in this field, the port number to be used is the default number 443. VMware User Name (Optional) The user name required for connecting to the VMware. VMware Password (Optional) The password of the VMware user specified above. 18 Managing Oracle Database Systems: User and Reference Guide 19 Chapter 2—Using Foglight for Oracle Configuring Multiple Instances for ASM To add multiple agents using the silent installation through a command line interface: 1 In the navigation panel, under Dashboards, click Administration > Cartridges > Components for Download. 2 Click the Installer icon beside DB_Oracle_ASM_CLI_Installer. 3 Follow your browser’s prompts to save the ZIP file. 4 Extract the ZIP file to a directory of your choice. The ZIP file contains the following files: • README.txt file • asm_cli_installer.groovy — a groovy script file that runs the silent installation. This file should be copied to the /bin directory. • asm_silent_installer_input_template.csv — a template file that should serve as the basis for inserting contents into the input CSV file. This file can be copied to any folder of your choice, provided that the path indicated by the parameter (see below) points to the selected path. For details about the contents of this file, see “Contents of the Input CSV File” on page 16 5 Copy the asm_cli_installer.groovy file to the /bin directory. 6 Go to the command line and execute the command: /bin/fglcmd -srv -port -usr pwd -cmd script:run -f asm_cli_installer.groovy fglam_name asm_instances_file_name lockbox_name lockbox_password The descriptions of the flags and parameters in this command are as follows: • • • • • • : The Foglight Management Server installation directory. : The name of the host where the Management Server is installed. : The Management Server HTTP port. : The user name used for connecting to the Management Server. : The password of the specified user. : The name of the selected Foglight Agent Manager where new agents are to be added. • : The full path and the name of the input CSV file. • : Optional - specifies the name of an existing lockbox to be used. • : Optional - the selected lockbox's password 7 After the installation process completes, review the CSV files that are created in the same folder where the input file resides: • A file with the input file's name and _status suffix (for example: if the input file is named input, this file is named input_status). Managing Oracle Database Systems: User and Reference Guide 20 Chapter 2—Using Foglight for Oracle This file indicates the monitoring situation of all Oracle instances listed in the input CSV file. If monitoring failed for one or more of the instances, the file specifies the error message. The _status file includes the name of the monitored Foglight for Oracle agent, the result of the monitoring validation process - MONITORED, FAILED, or AGENT EXISTS - and the error message, in case the agent creation failed. Important If all instances listed in the input CSV file passed the monitoring validation successfully, no additional file is created. • A file with input file's name and _new suffix (for example: if the input file is named input, this file is named input_new). This file, which is only created if one or more instances, nodes or RACs failed the monitoring validation, contains all details about these instances, RACs, or nodes. Using this CSV file, the user only needs to fix the errors and re-run the script, this time by specifying this file as the input file, that is: the original file name with suffix _new. In the example used here, the new file should be named input_new.csv. Contents of the Input CSV File The input CSV file contains the following fields, which are used as columns in the resulting file: Name Description RAC Agent Name The name of the RAC agent. This name must be unique per agent. • If the instance belongs to a RAC, this field has to contain the name of the RAC agent. • If this RAC runs multiple nodes, all instances must have the same RAC agent name. • If the instance does not belong to a RAC, leave this field empty. Agent Name The requested agent name. This name must be unique per agent. If an agent with the given name already exists, the newly provided agent will not be created. Host The host that is serving the ASM instance. Port The TNS listener port. SID The name of the SID used for establishing connection to the instance. Note User name The SID name must be preceded by \\. For example: \\+ASM1. The name of the user to be used for connecting to the ASM instance. Can be either ASMSNMP or SYS. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Name Description User Type Connection privileges level to be used for monitoring the ASM instance. The possible values are either "SYSASM" or "SYSDBA". Password The password of the database user name. Enable OS Monitoring Specifies whether to create an Infrastructure Cartridge agent for the given host name; the possible values are True or False. OS Platform (Optional) The type of the OS platform where the Oracle instance runs; the possible values are either UNIX or WINDOWS. Authentication (Optional) OS authentication type. This field can have one of the following values: • LOGIN_PASSWORD: user/password authentication • RSA_KEY: RSA-based public key authentication • DSA_KEY: DSA-based public key authentication • CLIENT_LOGIN: uses the Foglight Agent Manager's login • WINDOWS_ACCOUNT: Windows authentication, in the domain\user and password convention OS User Name (Optional) The user name to be used for connecting to the OS and monitoring it. OS Password (Optional) The password of the OS user name specified above. SSH Port (Optional) The port that is used for listening to incoming SSH connections on the monitored hosts; if the monitored host is Windows-based, leave this field empty. Private Key File Path (Optional) The full path of the private key file; relevant only to the following credential types: RSA_KEY, DSA_KEY. Passphrase (Optional) The passphrase of the OS user that is used in public key authentication; relevant only to the following credential types: RSA_KEY, DSA_KEY. Use sudo (Optional) Used for indicating that the OS user specified above is not an administrator user, but can run certain commands that require administrative privileges as root. The possible values are TRUE or FALSE. if the monitored host is Windows-based, leave this field empty to indicate FALSE. Reviewing the Foglight for Oracle Architecture The communication architecture of Foglight for Oracle is displayed below: 21 Managing Oracle Database Systems: User and Reference Guide 22 Chapter 2—Using Foglight for Oracle Configuring RMI Server on the Agent Manager Concentrator If a firewall is installed between the Management Server and the Agent Manager, the bidirectional RMI connection shown in the figure above will not be possible because RMI requests require direct connection. This issue can be resolved by creating an RMI proxy server on the concentrator — an Agent Manager instance that works similarly to an HTTP proxy. This RMI proxy server can be configured to accept RMI requests from the Management Server and direct them to the Agent Manager instances (called downstream instances). For additional details about the concentrator, see Configuring an Agent Manager Instance as a Concentrator in the Agent Manager Installation Guide. Managing Oracle Database Systems: User and Reference Guide 23 Chapter 2—Using Foglight for Oracle A concentrator agent, created specifically for Foglight for Oracle, allows controlling the RMI server so that the server is initialized (starts collecting data) when the agent starts, and stops when data retrieval by the agent is brought to an end. The agent, which is created manually by the user, requires configuring the following parameters: • A list of the downstream hosts that are connected through the concentrator • The RMI port number Each time the Foglight Management Server sends an RMI request, the list of downstream hosts submitted by the concentrator agent is being searched. • Inclusion of the target RMI server host name within this list indicates that a Foglight Agent Manager concentrator exists, in which case a connection is made to the proxy RMI server. • If the list does not contain the target host name, the connection is made directly to the target RMI server. To create the concentrator agent: 1 Go to Dashboards > Administration > Agents > Agent Status. 2 Click Create Agent. 3 Select the agent type DB_Oracle_Concentrator. 4 Enter a name of your choice in the Instance Name field. 5 Click Create. 6 After the creation process is completed successfully, click OK. To edit the concentrator agent connection details: 1 Select the newly created concentrator agent. 2 Click Edit Properties. 3 Click Modify properties for this agent only. 4 Enter the number of the proxy RMI server port, or accept the default number. 5 Enter the concentrator Foglight Agent Manager host name (optional). 6 Click Edit to edit the list of downstream Agent Managers. 7 Click Add Row. Managing Oracle Database Systems: User and Reference Guide 24 Chapter 2—Using Foglight for Oracle 8 Enter the name of the requested Foglight Agent Manager host. Important When editing the properties of the Foglight Agent Manager concentrator (Dashboards > Administration > Agents > Agent Status), the name of the Foglight Agent Manager should be entered in the Downstream FglAMs section exactly as it appears in the topology, under Home > Agents > All agents > > RMI data > Fglam host property. The parameter refers to any of the Foglight for Oracle agents that reside on the selected Foglight Agent Manager. 9 Repeat step 7 to step 8 as many times as required. 10 Click Save Changes. 11 Click Save at the lower right corner of the screen. 12 Click Back to Agent Status. 13 Select the newly created concentrator agent. 14 Click Activate. 15 Click OK to complete the process. Important In an environment that includes a Foglight Agent Manager concentrator, when upgrading the environment where the concentrator does not contain any regular agents, the concentrator is not upgraded. Therefore, after the upgrade process takes place the Foglight Agent Manager concentrator has to be deployed manually. Running the Oracle Monitoring Wizard This section provides the procedure to run the Oracle monitoring wizard, which offers several methods for adding Foglight for Oracle database instances. This procedure is detailed starting from section “Configuring Oracle Monitoring using the Monitoring Wizard” on page 25. In addition, this section also describes how to monitor Oracle in a clustered environment and how to prepare a sudo environment, which is used by UNIX and Linux users. Important The database addition process can be carried out to monitor only Oracle versions 9, 10, and 11 Monitoring Oracle in an OS Clustered Environment Foglight for Oracle can be used for remote monitoring of Oracle instances that reside on a cluster environment. To enable remote monitoring in an OS clustered environment: 1 Create a regular Oracle agent, using the Oracle Monitoring Installer wizard. Note Monitoring a RAC requires configuring the agents of the Cluster and all of its nodes to run under the same Foglight Agent Manager 2 When required to specify a host name, enter the cluster virtual host name. Foglight for Oracle will trigger an alarm when a cluster failover is detected, indicating the physical name of the cluster node that failed over. Managing Oracle Database Systems: User and Reference Guide 25 Chapter 2—Using Foglight for Oracle Preparing a sudo Environment for UNIX and Linux Users Foglight for Oracle supports using sudo on the various Linux and UNIX operating systems where it runs. The binary files that should be included per each of Foglight for Oracle’s supported platforms in order to enable using sudo are as follows: • HP-UX • Execute permissions: bdf uptime sar netstat vmstat ps hostname awk /usr/sbin/ ioscan uname getconf iostat lsnrctl • Read permissions: /var/adm/syslog/syslog.log • AIX • Execute permissions: df iostat uptime netstat vmstat ps hostname awk lsdev lsattr oslevel pagesize uname lsnrctl • Solaris • Execute permissions: df iostat uptime netstat vmstat ps hostname awk mpstat / usr/sbin/prtconf uname pagesize lsnrctl • Linux • Execute Permissions: df iostat uptime sed ps hostname awk free uname /proc/ vmstat vmstat netstat getconf lsnrctl • Read Permissions: /proc/net/dev /proc/stat /proc/vmstat on Linux >= 2.6 / proc/cpuinfo getconf Configuring Oracle Monitoring using the Monitoring Wizard Important The instructions below refer to the addition of Oracle instances and RACs. Instructions about the addition of other instance types are found in the following sections: “Adding ASM Instances” on page 34 “Adding CRS Agents for Monitoring Clusterware” on page 39 “Adding Agents for Monitoring Exadata Environments” on page 42 To configure instances for monitoring using the wizard: 1 Navigate to Homes > Databases. • If no instances are monitored, the Database view appears empty and you are prompted to click the button Monitor More Instances. • If one or more instances are already monitored, the Databases view displays a list of the currently monitored instances. Important If several database cartridges are installed, the Databases view displays all database types that can be monitored, even if no instance of a specific database type is currently monitored. In such a case, the instance count for this database type displays the number 0. If only one database cartridge is installed, the Databases view displays only this database type, without the All Databases tile. Managing Oracle Database Systems: User and Reference Guide 26 Chapter 2—Using Foglight for Oracle 2 Click Monitor More Instances at the upper left corner of the view. Important If a user-defined database group is currently selected, the databases table's title displays the name of this group instead of All; however, all newly added or created databases are added to the general (All) group of databases. If only one database cartridge is installed, the Oracle Monitoring Installer wizard view appears immediately. The screen that appears now allows selecting between the following options for adding more instances for monitoring: • Add one — suitable for adding a small number of instances, using default settings. For details, see “Adding a Single Oracle Instance” on page 31. • Discover one or more — suitable for medium and large environments with complex configurations and deployments, possibly requiring user-defined settings. For details. see “Configuring Oracle Monitoring using the Monitoring Wizard” on page 25. Important Configuring the Foglight Performance Analysis connectivity is not carried out using the instance installation wizard. If one or more of the newly added instances has Foglight Performance Analysis installed and configured, the Foglight Performance Analysis connectivity can be configured only using the Connection Details view in the Databases Administration dashboard. For details, see “Configuring user-defined Performance Analysis connectivity details” on page 170. Adding one or more databases using the Oracle monitoring wizard 1 Select the option Discover one or more. 2 Click Oracle. The Introduction screen appears, displaying the monitoring architecture. Managing Oracle Database Systems: User and Reference Guide 27 Chapter 2—Using Foglight for Oracle 3 Click Next. 4 Use the Select Foglight Agent Manager screen to determine whether to select the host recommended by the wizard (the host that contains the smallest number of monitored instances) or a different host. 5 Click Next. 6 Use the following screen to add instances for monitoring, by clicking Add instances. Note Ensure that the login credentials and instance details (instance name and OS) for all Oracle instances to be monitored are available. The methods available for adding instances are loading a TNS file or adding a single Oracle instance. To add a single Oracle instance, go to section “Adding a Single Oracle Instance” on page 31; otherwise proceed with the steps below. 7 To load a TNS file: a Click Load TNS file. The Select a file dialog box opens. b Click Choose file and browse for the requested file. c Click Open. d Click OK. The Oracle Instances dialog box appears, displaying the Oracle instances that were discovered in the provided tnsnames file. Note The Database list only displays instances that are not yet monitored. e Select the check boxes beside the requested instances and click Select. The selected instances are now displayed on the table with the status Set credentials. Managing Oracle Database Systems: User and Reference Guide 28 Chapter 2—Using Foglight for Oracle The Set credentials button The Set credentials link To set the credentials of a specific instance, click Set credentials in the Status column of the requested agent and proceed to step 10. To set credentials for multiple instances, proceed to step 8. Important Clicking Set credentials on the Status column of a specific instance leads to the exclusive selection of this instance, regardless of the number of instances whose check box are selected. 8 Select the check boxes beside the instances whose credentials are to be configured. 9 Click Set credentials on the toolbar to provide the login credentials of the instances. The dialog box used for editing the credentials of the selected instance, or all instances selected, opens. 10 Fill in the following details: Managing Oracle Database Systems: User and Reference Guide 29 Chapter 2—Using Foglight for Oracle a Use the Login Credentials section to specify the username and password to be used for monitoring the Oracle instance. b Select whether to enable OS monitoring, and specify the requested authentication type: • Local user — using the same credentials that were used to launch the Agent Manager. • Windows — the user name (in the domain\user name format) and password, used for logging in to a Windows-based monitored host. • SSH (DSA) — using the SSH private key, generated via the DSA algorithm. For more information, see “Enabling OS monitoring Using a DSA Key” on page 33. • SSH (RSA) — using the SSH private key, generated via the RSA algorithm. • SSH (login credentials) — the user name and password used for logging in to a UNIXbased monitored host. c To select existing credentials using the Foglight lockbox mechanism, click Select from stored credentials... The Stored Credentials dialog box that appears now allows reviewing the login credentials and authentication methods used for logging in to Foglight. Foglight stores encrypted credentials in lockboxes, which may be password-protected for added security. The default lockbox is DB-Agent Lockbox. If credentials have already been entered in another lockbox, use the Lockbox list to select from that lockbox. After selecting the required user, click Apply to return to the Edit Credentials dialog box If the requested Oracle instance runs on a virtual machine, go to step 11. Otherwise, skip to step 12. 11 If the selected instances reside on a virtual machine, click VMWare connection details to configure the connection details to the vCenter server or the ESX server that hosts the virtual machine. The VMware Connection Details dialog box opens. Use this dialog box to select whether to enable monitoring the VMware layer. Selecting the check box that enables the monitoring allows providing the following details: • The name or IP address of the vCenter server that hosts the Oracle instance virtual machine, or the name of its parent ESX server. Managing Oracle Database Systems: User and Reference Guide 30 Chapter 2—Using Foglight for Oracle • The port number used by the vCenter server system or by the ESX server for listening to the connections from the vSphere Client (default: 43). • The name of the user that has the privileges required for connecting to the vCenter server or ESX server and retrieving information. • The user password. Click OK to apply these changes. 12 Click Validate. If during the connectivity validation process several RACs with a single active node are detected, a dialog box appears, asking to identify the RACs that are not RAC One Node. Select the check boxes near the multi-node RACs and click OK. After the validation process completes, all instances that passed the validation process successfully are displayed in the table with the status Validated. However, the validation can fail for several reasons. To view these reasons, go to step 10 on page 32. Important Adding a node to an Oracle RAC, as well as configuring the PA connectivity, is only possible using the Connection Details view on the Databases Administration dashboard. For details, see “Defining Connection Details” on page 166 and “Configuring the Connection to Performance Analysis” on page 193. 13 Select the validated instances and click Next. 14 Use the last screen, Monitor Oracle Instances, to review the table that displays all instances that passed the connectivity validation successfully. • To add other instances or modify the connection details of existing instances, click Previous. • To confirm the selection and start monitoring the instances, click Next. After clicking Next, the table displays all instances that were configured for monitoring. Managing Oracle Database Systems: User and Reference Guide 31 Chapter 2—Using Foglight for Oracle Adding a Single Oracle Instance To add a single Oracle instance: 1 Click Monitor more instances. 2 Click Oracle under the Add one option. 3 If more than one Agent Manager suitable for monitoring your Oracle instance is detected, a dialog box notifying this fact appears. This dialog box allows choosing whether to proceed with the agent that was found to be the optimal Agent Manager to be used or to click a link that leads to the Discover More Instances wizard, described in section “Configuring Oracle Monitoring using the Monitoring Wizard” on page 25. 4 Click OK. The Monitor New Oracle Instance dialog box opens. Managing Oracle Database Systems: User and Reference Guide 32 Chapter 2—Using Foglight for Oracle 5 Use the Connection Details section to select the requested method for entering connection details: • TNS — used for selecting a database from the tnsnames.ora file by clicking Load file. Note The Database list only displays instances that are not yet monitored. • Manual — allows typing the host, port, and either service name or SID. 6 Use the Login Credentials section to specify the username and password to be used for monitoring the Oracle instance. 7 Select whether to enable OS monitoring, and specify the requested authentication type: • Local user — using the same credentials that were used to launch the Agent Manager. • Windows — the user name (in the domain\user name format) and password, used for logging in to a Windows-based monitored host. • SSH (DSA) — using the SSH private key, generated via the DSA algorithm. For more information, see “Enabling OS monitoring Using a DSA Key” on page 33. • SSH (RSA) — using the SSH private key, generated via the RSA algorithm. • SSH (login credentials) — the user name and password used for logging in to a UNIXbased monitored host. 8 Enter the requested user name, password, and port. 9 To use Sudo, select the Sudo check box. 10 Click Monitor. If the credential validation completed successfully, the newly configured instance will be monitored. However, the validation can fail for various reasons, such as invalid user name or password. Some of the most common reasons for failure are as follows: • Insufficient database privileges • The Selected Agent Manager has reached its full monitoring capacity. For details, go to “Handling overloaded Agent Managers” on page 33. If the credentials provided for the Oracle database have insufficient database privileges for accessing Oracle data dictionary views, click the status Insufficient privileges to open the Grant Database Privileges dialog box, which provides the following methods for granting database privileges: • Enter a SYSDBA user and password. • Grant privileges manually, by clicking View grant script. To view the script used for granting the privileges, click View grant script. Important The script for granting privileges is invoked using a popup. To view the script, ensure that pop ups are not blocked on the page. Managing Oracle Database Systems: User and Reference Guide 33 Chapter 2—Using Foglight for Oracle 11 Click Grant. Handling overloaded Agent Managers If the Agent Manager has reached its monitoring capacity, the Full Monitoring Capacity Reached dialog box opens. It is highly recommended to read the Hardware Sizing and Tuning Guide in order to view the settings defined for the monitoring capacity of agent managers. To do that, click For more information. Important Because this link leads to the web page containing the Hardware Sizing and Tuning Guide, the link is treated as a popup. To view the guide, ensure that pop ups are not blocked on the page. Enabling OS monitoring Using a DSA Key Starting from version 5.6.4.330, OS monitoring is also enabled using a DSA key. To enable OS monitoring using a DSA key (UNIX users only): 1 Create one or more Oracle agents using the Oracle Monitoring Installer wizard (without OS monitoring). Important Disabling OS monitoring in the Oracle Monitoring Installer wizard prevents the wizard from automatically creating the Infrastructure cartridge agent, without which several OS metrics cannot be retrieved. However, this agent can be created manually using the Managing Oracle Database Systems: User and Reference Guide 34 Chapter 2—Using Foglight for Oracle Dashboards > Hosts > Hosts dashboard. For details, see “Adding a Monitored Host” in the Managing the Infrastructure Cartridge User and Reference Guide. 2 Navigate to the Foglight Management Server credentials view (Dashboards > Administration > Cartridges > Credentials > Manage Credentials). 3 Add credential of type DSA key and link this credential to the agents created in step 1. For details, see “Adding and deleting credentials” in the Foglight Administration and Configuration Guide. 4 Go to Agent Status screen (Administration > Agents > Agent Status). 5 Select the agent or agents created in step 1. 6 Click Edit Properties. 7 Select Modify the private properties for this agent. 8 Under the Agent Management section, set Enable OS collections to True. Foglight-Specific Views Foglight retrieves metrics for space fragmentation, initialization parameters, and backup jobs from the Oracle X$% objects for display in various dashboards. Because only users with a SYSDBA authorization can directly run queries against Oracle's X$% objects, Foglight creates its own views for the required queries and grants select on QUEST views to the foglight user used for monitoring. In so doing, Foglight helps keep the privilege requirements for the foglight user minimal, thereby denying the user system-related privileges. The Foglight standard is to use an X_QUEST prefix in the name of all newly created views. Specifically, Foglight creates the views listed below. • X_QUEST_$RMAN_STATUS_CURRENT • X_QUEST_$KSPPCV • X_QUEST_$KSPPI • X_QUEST_$KTFBUE • X_QUEST_$KTFBFE A Foglight deployment does not create any objects that write data to SYS schema or anywhere else within the database; Foglight objects on the database are limited to views, synonyms, and so on. As a result, these objects do not alter the IO performance characteristics of any tablespace. Adding ASM Instances The process of adding ASM instances is almost identical to the process of adding Oracle instances. To add ASM instances: 1 Go to the ASM tab. 2 Click Monitor more ASM instances at the upper left corner of the Databases table. The Introduction screen appears, displaying the monitoring architecture. Managing Oracle Database Systems: User and Reference Guide 35 Chapter 2—Using Foglight for Oracle 3 Click Next. 4 Use the Select Foglight Agent Manager screen to determine whether to select the host recommended by the wizard (the host that contains the smallest number of monitored instances) or a different host. 5 Use the following screen to add ASM instances for monitoring, by clicking Add instances. Note Ensure that the login credentials and instance details (instance name and OS) for all ASM instances to be monitored are available. The methods available for adding instances are loading a TNS file or adding a single ASM instance. To add a single ASM instance, go to section “Adding a Single ASM Instance for Monitoring” on page 37; otherwise proceed with the steps below. 6 To load a TNS file: a Click Load TNS file. The Select a file dialog box opens. b Click Choose file and browse for the requested file. c Click Open. d Click OK. The ASM Instances dialog box appears, displaying the ASM instances that were discovered in the provided tnsnames file. Note The Database list only displays instances that are not yet monitored. e Select the check boxes beside the requested instances and click Select. The selected instances are now displayed on the table with the status Set credentials. Managing Oracle Database Systems: User and Reference Guide 36 Chapter 2—Using Foglight for Oracle The Set credentials button The Set credentials link To set the credentials of a specific instance, click Set credentials in the Status column of the requested agent and proceed to step 10. To set credentials for multiple instances, proceed to step 8. Important Clicking Set credentials on the Status column of a specific instance leads to the exclusive selection of this instance, regardless of the number of instances whose check box are selected. 7 Select the check boxes beside the instances whose credentials are to be configured. 8 Click Set credentials on the toolbar to provide the login credentials of the instances. The dialog box used for editing the credentials of the selected instance, or all instances selected, opens. Managing Oracle Database Systems: User and Reference Guide 37 Chapter 2—Using Foglight for Oracle 9 Fill in the following details: a Use the Login Credentials section to specify the username and password to be used for monitoring the ASM instance. b Select whether to connect as SYSDBA or SYSASM. Note SYSASM is supported only from Oracle 11g release 2. Select whether to enable OS monitoring. If OS does not need to be monitored, proceed to step 12; otherwise proceed to the next step. 10 If OS monitoring is to be enabled, specify the requested authentication type: • Local user — using the same credentials that were used to launch the Agent Manager. • Windows — the user name (in the domain\user name format) and password, used for logging in to a Windows-based monitored host. • SSH (DSA) — using the SSH private key, generated via the DSA algorithm. For more information, see “Enabling OS monitoring Using a DSA Key” on page 33. • SSH (RSA) — using the SSH private key, generated via the RSA algorithm. • SSH (login credentials) — the user name and password used for logging in to a UNIXbased monitored host. 11 To select existing credentials using the Foglight lockbox mechanism, click Select from stored credentials... The Stored Credentials dialog box that appears now allows reviewing the login credentials and authentication methods used for logging in to Foglight. Foglight stores encrypted credentials in lockboxes, which may be password-protected for added security. The default lockbox is DB-Agent Lockbox. If credentials have already been entered in another lockbox, use the Lockbox list to select from that lockbox. After selecting the required user, click Apply to return to the Edit Credentials dialog box. 12 Click Validate. After the validation process completes, all instances that passed the validation process successfully are displayed in the table with the status Validated. 13 Select the validated instances and click Next. 14 Use the last screen, Monitor ASM Instances, to review the table that displays all instances that passed the connectivity validation successfully. • To add other instances or modify the connection details of existing instances, click Previous. • To confirm the selection and start monitoring the instances, click Next. After clicking Next, the table displays all instances that were configured for monitoring. Adding a Single ASM Instance for Monitoring To add a single ASM instance: 1 Click Monitor more ASM instances. 2 Click Add one. Managing Oracle Database Systems: User and Reference Guide 38 Chapter 2—Using Foglight for Oracle 3 If more than one Agent Manager suitable for monitoring your Oracle instance is detected, a dialog box notifying this fact appears. This dialog box allows choosing whether to proceed with the agent that was found to be the optimal Agent Manager to be used or to click a link that leads to the Discover More Instances wizard, described in section “Adding ASM Instances” on page 34. 4 Click OK. The Monitor New ASM Instance dialog box opens. 5 Use the Connection Details section to select the requested method for entering connection details: • TNS — used for selecting a database from the tnsnames.ora file by clicking Load file. Note The Database list only displays instances that are not yet monitored. • Manual — allows typing the host, port, and either service name or SID. 6 Use the Login Credentials section to specify the username and password to be used for monitoring the ASM instance. 7 Select whether to connect as SYSDBA or SYSASM (Oracle version 11G release 2 and above). 8 Select whether to enable OS monitoring. Select whether to enable OS monitoring. If OS does not need to be monitored, proceed to step 11; otherwise proceed to the next step. 9 If OS monitoring is to be enabled, specify the requested authentication type: Managing Oracle Database Systems: User and Reference Guide 39 Chapter 2—Using Foglight for Oracle • Local user — using the same credentials that were used to launch the Agent Manager. • Windows — the user name (in the domain\user name format) and password, used for logging in to a Windows-based monitored host. • SSH (DSA) — using the SSH private key, generated via the DSA algorithm. For more information, see “Enabling OS monitoring Using a DSA Key” on page 33. • SSH (RSA) — using the SSH private key, generated via the RSA algorithm. • SSH (login credentials) — the user name and password used for logging in to a UNIXbased monitored host. 10 To use Sudo, select the Sudo check box. 11 Click Monitor. If the installer detected that the ASM instance is part of an ASM cluster, a dialog box appears, asking to verify the connection details. 12 Review all details and click Monitor. Adding CRS Agents for Monitoring Clusterware Receiving alarms about the CRS software, which is used for monitoring Oracle Clusterware, requires creating CRS agents. Note The Monitor More Clusterwares wizard supports the monitoring of a single CRS cluster. Configuring multiple CRS clusters for monitoring requires re-running the wizard for each cluster. To add agents for monitoring Clusterware: 1 Click the Oracle tile. 2 Go to the CRS tab. 3 Click Monitor more Clusterwares at the upper left corner above of the Databases table. The Monitor More Clusterwares wizard opens. 4 In the Configure CRS Agent screen, use the CRS agent name field to set a user-defined name for the agent. Alternatively, accept the default name provided for the CRS agent. 5 Select the requested platform for the CRS nodes: either UNIX or Windows. 6 Click Next. Managing Oracle Database Systems: User and Reference Guide 40 Chapter 2—Using Foglight for Oracle 7 Use the Configure Nodes for Monitoring screen to add the requested CRS nodes, using either of the following methods: • Adding the nodes manually — go to step 14 on page 40. Note Adding the nodes manually requires specifying the CRS_HOME. • Discovering the nodes — proceed to step 8. 8 Click Add Nodes, and from the menu select Discover by cluster node. The Discover Nodes dialog box opens. 9 Use the Node Details section to enter the following details: • Node name — the name of one of the nodes, through which all other nodes in the cluster are to be discovered. • CRS_HOME (optional) — the path to the CRS Home directory. If this field is left empty, the installer automatically detects the path (recommended). 10 Use the OS Credentials section to enter the OS credentials of the requested user. Note Ensure providing a user that has permissions to execute CRS utilities. a To select existing credentials using the Foglight lockbox mechanism, click Select from stored credentials... The Stored Credentials dialog box that appears now allows reviewing the login credentials and authentication methods used for logging in to Foglight. b Foglight stores encrypted credentials in lockboxes, which may be password-protected for added security. The default lockbox is DB-Agent Lockbox. If credentials have already been entered in another lockbox, use the Lockbox list to select from that lockbox 11 Fill in the other details (optional) and click Apply. 12 The wizard tries to connect to the node specified, and through this node to all other nodes that take part in the CRS cluster. All nodes are validated for OS connectivity and the permissions of the provided user are verified. After this process completes, nodes that were successfully validated appear in the table with the status Connection verified, while other nodes display the relevant error message (for example: Failed to establish connection to host). 13 Click the error message to view the error. To handle the issues indicated in the message, click Set credentials, enter the correct details and click Apply, then proceed to step 19 on page 41. Note If the error results from specifying the wrong host, click Remove and repeat the node addition process. 14 Start the process of manually adding cluster nodes for monitoring by entering the CRS_HOME of the requested cluster. Click the CRS_HOME link at the bottom of the screen. Managing Oracle Database Systems: User and Reference Guide 41 Chapter 2—Using Foglight for Oracle CRS_HOME link The Edit CRS Home dialog box opens. 15 Enter the path of the requested CRS_HOME, then click OK. 16 Click Add nodes. 17 Select the option Add manually. The Add CRS Nodes dialog box opens. This dialog box is similar to the Discover Nodes dialog box, with the exception of the CRS_HOME parameter that has already been configured. 18 Follow step 9 to step 13, then proceed to step 19. 19 Click Next. The agent is now created, based only on nodes that have been validated. After the agent’s creation completes, monitoring starts automatically, and the Summary screen appears, displaying the newly created agent with all its components. 20 Click Finish to return to the Databases dashboard. The CRS tab now displays data about all newly monitored nodes. Modifying the Clusterware Monitoring Settings After the CRS agents have been configured for monitoring, their monitoring settings can be modified using the Agent Settings button. Important Unlike Foglight for Oracle instances, the agent properties of CRS agents cannot be modified using the Databases Administration dashboard, but only through the default Foglight tools (see Foglight Administration and Configuration Guide). Clicking Agent settings and selecting the Administration option only allows re-running the wizard used for configuring the instances for monitoring, and changing the options defined using this wizard. To modify the CRS monitoring settings: 1 Select the requested CRS agents. 2 Click Agent settings. Managing Oracle Database Systems: User and Reference Guide 42 Chapter 2—Using Foglight for Oracle 3 Select Administration to re-run the wizard used for configuring the agents for monitoring, and change the options defined using this wizard. 4 Select the requested option for changing the agent’s monitoring: • Start monitoring • Stop monitoring • Remove from monitoring Adding Agents for Monitoring Exadata Environments To add agents for monitoring Exadata environments: 1 Click the Oracle tile. 2 Go to the Exadata tab. 3 Click Monitor more Exadata environments at the upper left corner above of the Databases table. The Monitor More Exadata Environments wizard opens. On the first screen of this wizard, Monitor Exadata, use the Exadata agent name field to set a user-defined name for the agent. Alternatively, accept the default name provided for the Exadata agent. 4 Click Next. Use the next screen, Monitor InifiniBand Switches, to configure the InfiniBand switches for monitoring, either by adding the switches manually or by discovering by switch name. Configuring the switches for monitoring allows retrieving InfiniBand topology information, as well as automatically detecting the Exadata cells. Note This screen is optional; if you prefer not to monitor the InfiniBand switches, you can skip this stage by clicking Next and proceeding directly to configure the Exadata cells for monitoring. In such a case, go to step 13. 5 Click Add switches. 6 From the menu that appears now, choose the requested option: • Add manually — go to step 7 on page 42. • Discover by switch name — go to step 9 on page 43. 7 Select the option Add manually. The Add InfiniBand Switch dialog box opens. Use this dialog box to add all requested InfiniBand switches that are monitored by the same credentials (user name and password), by entering a comma-separated list of all the names of the hosts on which these switches reside. To add switches that would be monitored by a different credentials set, click Apply and repeat the process with the other credentials. To edit the credentials of an existing switch, use either of the following methods: • For a single switch, click Set credentials on the switch’s Status row. Managing Oracle Database Systems: User and Reference Guide 43 Chapter 2—Using Foglight for Oracle • For multiple switches, select the check boxes near the requested switch and click Set credentials. The Set credentials dialog box opens. This dialog box is identical to the Add InfiniBand Switch dialog box, except for the host name that cannot be modified after it was set. To edit a single switch name, click the Switch name column. In the Edit Switch Name dialog box that appears, use the Switch name field to enter a new name for the switch. Note The switch name specified must be taken from the list displayed by running the ibswitches command. 8 After adding all requested switches, click Next to proceed to step 12. 9 Select the option Discover by switch name. The Discover InfiniBand Switches and Cells dialog box opens. 10 Use this dialog box to enter the host name and login credentials of a single InfiniBand switch, to be used for the discovery of all switches and cells. 11 After adding all requested details, click Apply and then click Next to proceed to step 12. 12 The following screen, Monitor Exadata Cells, contains a table that lists all of the Exadata Cells that were automatically discovered when the InfiniBand switches were configured for monitoring. If the stage of configuring InfiniBand switches was skipped, this table is empty and cells need to be added manually. 13 Click Add. The Add Cells dialog box opens. Use this dialog box to add all of the requested Exadata cells that are monitored by the same credentials (user name and password), by entering a comma-separated list of all the names of the hosts on which these cells reside. To add cells that would be monitored by a different credentials set, click Apply and repeat the process with the other credentials. To edit the credentials of an existing cell, use either of the following methods: • For a single cell, click anywhere on the switch’s Status row. • For multiple cells, select the check boxes near the requested cells and click Set credentials. The Set credentials dialog box opens. This dialog box is identical to the Add Cells dialog box, except for the host name that cannot be modified after it was set. 14 Click Apply. After the connectivity and credentials were successfully validated, all cells appear with the status Connection Verified. 15 Click Next. The agent is now created, based only on switches and cells that have been validated. After the agent’s creation completes, monitoring starts automatically, and the Summary screen appears, displaying the newly created agent with all its components. 16 Click Finish to return to the Databases dashboard. Managing Oracle Database Systems: User and Reference Guide 44 Chapter 2—Using Foglight for Oracle The Exadata tab now displays all newly monitored cells, along with partial monitoring data. To view in-depth data about the Exadata environment, click the Home page button on the Exadata tab to access the Exadata drilldown. For details, see “Reviewing Exadata-related Information” on page 157. Modifying the Exadata Monitoring Settings After the Exadata agents have been configured for monitoring, their monitoring settings can be modified using the Agent Settings button. Important Unlike Foglight for Oracle instances, the agent properties of ASM instances cannot be modified using the Databases Administration dashboard, but only through the default Foglight tools (see Foglight Administration and Configuration Guide). Clicking Agent settings and selecting the Administration option only allows re-running the wizard used for configuring the agents for monitoring, and changing the options defined using this wizard. To modify the Exadata monitoring settings: 1 Select the requested Exadata switch. 2 Click Agent settings. 3 Select Administration to re-run the wizard used for configuring the agents for monitoring, and change the options defined using this wizard. 4 Select the requested option for changing the switch’s monitoring: • Start monitoring • Stop monitoring • Remove from monitoring Introducing the Foglight for Oracle Drilldowns This chapter describes the drilldowns accessible through the Foglight for Oracle dashboard. Overview Foglight for Oracle includes a home page that provides an overall view of the status of the selected entity (either Oracle RAC or Oracle instance) for the selected time range, as well as drilldowns, which provide a visual representation of the status and performance of the monitored Oracle environment. Use these drilldowns to determine any potential bottlenecks in database performance. Managing Oracle Database Systems: User and Reference Guide 45 Chapter 2—Using Foglight for Oracle The Foglight for Oracle Layout Oracle instances and clusters are listed in the Database table of the Databases dashboard, accessible by clicking Dashboards > Databases > Databases on the navigation panel. Note The user privileges that affect dashboard access are granted when running the Oracle Monitoring Installer wizard. The following diagram provides an overview of the Foglight for Oracle navigation. Note In RACs where Exadata is configured, the Exadata drilldown is added in both Whole cluster and Instance (single node) views, when the ASM performance is monitored through Exadata. Note The Pluggable Databases drilldown appears only when Foglight for Oracle is monitoring a multitenant architecture. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle 46 Managing Oracle Database Systems: User and Reference Guide 47 Chapter 2—Using Foglight for Oracle Reviewing the Databases Dashboard Note The following section describes the various components of the Databases dashboard. To learn how to customize this dashboard’s display and focus on a requested instance, see “Selecting the Instance to be Monitored” on page 54. The Foglight for Oracle Overview page is launched using the Databases dashboard. To access the Databases dashboard: • On the navigation panel, click Homes > Databases. Tip To open the navigation panel, click the right-facing arrow on the left. The Databases dashboard provides an at-a-glance view of the monitored environment, with all of the currently monitored database types. Status summary Instance installer Oracle instances and agents Database-specific configuration health summary Link to Databases Administration dashboard Go to Overview button Database cartridge type Select Dashboard button The Databases dashboard comprises the following sections: Currently selected database group table Managing Oracle Database Systems: User and Reference Guide 48 Chapter 2—Using Foglight for Oracle • Status section — includes the following components: • Status summary — a color-coded bar, which provides a visual representation of the summarized health condition of all databases listed in the Database Group table. The status summary bar provides a representation of the monitored environment’s current state, broken down to the number of instances and their current health state: Fatal, Critical, Warning, Normal, or Unknown. • Database-specific health summary — when the database group All is selected, this section displays all of the currently monitored instances for each database type, divided by their health state (for example: four Oracle instances, three of which have the health state Warning and one is indicated with the health state Fatal). When a user-defined database group is selected, this section displays data only about the agents included within the selected group. The Status Indicators section can be used for filtering the Databases Group table to display only instances that meet a criterion set in this section. For details, see “Filtering the Display by Severity” on page 55. • Database cartridge type tiles — each tile represents a database type (SQL Server, Oracle, Sybase, DB2, or All) and displays the number of instances for each database type, along with a breakdown according to the instance health state severity (Normal, Warning, Critical, or Fatal). • Oracle instances and advanced configuration — this section, which only appears if the Oracle database type is selected in a multi database environment, contains the following tabs: • Oracle Instances — used for monitoring Oracle Instances and RACs, as described in this chapter. • ASM — Starting from version 5.6.4.32x ASM-related data, which was previously retrieved through the Foglight for Oracle agent, is only enabled if the Foglight ASM instance is configured and running. For details about configuring and running this instance, see “Adding ASM Instances” on page 34. • CRS — used for configuring the monitoring of the Clusterware software that manages the oracle RACs. For details about configuring Clusterware for monitoring, see “Adding CRS Agents for Monitoring Clusterware” on page 39. • Exadata — used for configuring the monitoring of the Exadata environment. After the Exadata agents are added, data retrieved from Exadata is displayed on this tab, as well as on the Exadata drilldown. For details about configuring the Exadata environment for monitoring, see “Adding Agents for Monitoring Exadata Environments” on page 42. For details about the monitoring information displayed on the Exadata tab, see “Viewing Data about the Exadata Cells” on page 53. For details about the drilldown that displays data retrieved from Exadata, see “Reviewing Exadata-related Information” on page 157. • Advanced operations section — includes the following buttons: • Monitor more instances — click this button to add instances to be monitored. For details, see “Adding and Configuring Agents” on page 14. • Agent settings button — after selecting one or more instances of the same database type, click this button and select Administration to set options for collecting, storing, and Managing Oracle Database Systems: User and Reference Guide 49 Chapter 2—Using Foglight for Oracle displaying data for the selected instances. For details, see “Managing Foglight for Oracle Agent Settings” on page 164. Important While the CRS and Exadata tabs contain the Agent Setting button, the properties of these agents cannot be edited using the Databases Administration dashboard, but only through the default Foglight tools (see Foglight Administration and Configuration Guide). Selecting such agents and clicking Agent settings re-runs the wizard used for configuring the agents, allowing you to change only monitoring-related properties defined in the wizard. The other commands available by clicking Agent Settings allow modifying the agent’s monitoring state, as follows: • Start monitoring • Stop monitoring • Remove from monitoring • Currently selected database group table — a list of all monitored databases within the database group that is currently selected in the Databases section. For details, see “Using the Currently Selected Database Group Table” on page 51. • The Go to Overview icon — provides direct link to the Overview drilldown. • The Select dashboards icon — provides direct link to several drilldowns and panels, as shown below, thereby saving the need to navigate to these locations through the Overview drilldown. The Status Indicators section can be used for filtering the Databases Group table to display only databases that meet a criterion set in this section, that is, a certain type of severity of a specific indicator (for example: fatal severity of the Availability indicator). The selected filter is shown in the table’s title. To select another filter, click the required status indicator (for example: CPU > Normal severity). To display the entire list of databases, click Clear filters. Each row on this table displays a specific database instance or RAC, and is used for accessing the monitored entity’s home page (Overview) or other panels, using one of the following methods: • Clicking the Go to Overview icon • Clicking the Select dashboard icon and selecting the requested dashboard or panel. • Clicking anywhere on the database row. A quick view panel of the database appears, displaying the database’s principal data. Clicking the Overview button in this section opens the database’s home page (Summary panel), while clicking Performance opens the Performance drilldown. Managing Oracle Database Systems: User and Reference Guide 50 Chapter 2—Using Foglight for Oracle • The Databases section. The Databases section appears on the navigation panel, upon selecting Homes > Databases. This section includes the components listed below. • Database Groups — includes by default the All list, which contains all of the existing databases. In addition, this component contains the Database Groups list, which serves as the parent group for creating user-defined sub groups of database instances. Such sub groups are recognized as Foglight services. For details, see “Creating User-defined Database Groups” on page 55. The Database Groups section also displays the highest severity level of the database groups, under the column Sev. • Services — displays the configured services that contain at least one object (database topology) that is used by Foglight for Oracle. All Foglight for Oracle sub groups created using the Database Groups section are also displayed under Services. In addition, Services displays groupings of monitored resources that were created using the Service Builder dashboard, and contain Foglight for Oracle database topology. The Services section also contains the column SLC (Service Level Compliance), which indicates the current availability of the selected service over a given period of time. For details about the Service Builder dashboard, see Foglight Online Help. Components Shared by All Foglight for Oracle Screens The tables’ Customizer button and the In-context action button are common to all Foglight for Oracle screens, as described below. Managing Oracle Database Systems: User and Reference Guide 51 Chapter 2—Using Foglight for Oracle In-context actions button • The Customizer button — use this button, which is found on the upper right corner of each table, to create a custom filter for the table. The filter is created by clicking Customizer and specifying the criteria that the various values should meet in order to be displayed in the table (having a specific name, exceeding a certain size, and so on). For detailed instructions and examples, see Foglight User Guide > Working with Dashboards > Working with Tables. • In-context actions button — found on the upper right corner of all screens, and provides the following options: • Agent settings — when using panels whose configuration is set using the Databases Administration dashboard, clicking this button opens the relevant screen in the Databases Administration dashboard (for example, clicking this button from within the Databases > Locks panel opens the Locks screen). • Refresh — clicking this button retrieves the data acquired using the most recent sampling. • Useful links — used for carrying out further investigation using the following external sources: • Investigate in Foglight Performance Analysis — if Foglight Performance Analysis was installed and configured as part of the Foglight for Oracle installation, launch this application to utilize the enhanced performance monitoring capabilities it provides. • Navigate to Oracle Performance — a central knowledge base about all of Dell Software’s database monitoring products. • Contact support — central knowledge base about all of Dell Software’s products. Using the Currently Selected Database Group Table The currently selected database group table displays the columns listed below. Name Description Instance Sev The highest severity alarm of the Oracle database (whether a database instance, RAC or RAC node), which determines the database’s health state. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Name Description Name The name of the Oracle instance, RAC or RAC node. In addition, this row includes the Go to Home Page button. Version The version number of the Oracle database instance. Up Since The date and time that the Oracle database instance (or, in a RAC, the instance that was the earliest to restart) was last restarted. Workload The workload (average active sessions) for the Oracle database instance or RAC. When holding the cursor over the workload graph, the dwell displays a chart that shows the workload history over the specified time range. Alarms The number of warning, critical, and fatal alarms for the Oracle database instance/RAC. When holding the cursor over one of the alarm counts, the dwell displays the most recent alarms invoked for this database, sorted by severity. Clicking this field displays the Alarms list, which is listed by severity order. See the Foglight Online Help, Monitoring System-Wide Alarms for details on the alarm information. System Utilization Host The name of the computer that is hosting the database instance or RAC. CPU Load (%) The overall operating system CPU usage by all processes (including CPU usage by the database). Memory (%) The percentage, within the total memory, of memory consumed by all operating system processes (including the database). This value includes both RAM resident memory and swapped memory. Disk (% Busy) The percentage of time the busiest device spent serving system-wide input/ output requests. This metric serves as a measure for the system I/O load. Monitoring Status Agent The operational status of the monitoring agent. When the agent instance is running, the State icon ( ) is green, and holding the cursor over the icon displays the message Collecting Data. When the agent instance is running but not collecting data, holding the cursor over the icon displays one of the following status messages listed below. • Starting • Stopped • Stopping • Unknown 52 Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Name Description OS The status of the OS data retrieval. PA Note While the Foglight for Oracle agent retrieves OS data, several OS metrics cannot be retrieved without the Infrastructure cartridge agent. IF OS monitoring was disabled during the creation of the Foglight for Oracle agent through the Oracle Monitoring Installer wizard, the wizard cannot automatically create the Infrastructure cartridge agent; However, this agent can be created manually using the Dashboards > Hosts > Hosts dashboard. For details, see Adding a Monitored Host in the Managing the Infrastructure Cartridge User and Reference Guide. Note If the Infrastructure agent was created, but OS database is collected only by the Foglight for Oracle agent, click the OS status to OS data collection through the Infrastructure agent. The status of Foglight Performance Analysis; indicates whether it is installed and configured for this agent. Foglight Performance Analysis is available only if it is installed as part of the Foglight for Oracle installation. Note The Foglight Performance Analysis configuration indication also appears on the Foglight for Oracle’s home page. For details, see “Reviewing the Status of the Connection to Performance Analysis” on page 100. Viewing Data about the Exadata Cells The table displayed in the Exadata tab contains the columns listed below. Name Description Exadata Sev The highest severity alarm invoked within all cells monitored by the selected agent, which determines the agent’s health state. Name The name of the Exadata agent. Version The version number of the Oracle Linux used for running the Exadata storage server. Online Cells The number of cells monitored by the selected agent that are currently online. Temperature Status The temperature status of all cells monitored by the selected agent, as provided by the Baseboard Management Controller (BMC); indicates whether the temperature is normal. Any deviation from the normal temperature value, which is determined by Oracle, invokes an alarm indicated using an orange color on this column. 53 Managing Oracle Database Systems: User and Reference Guide 54 Chapter 2—Using Foglight for Oracle Name Description Total Read Rate (Mb/S) The rate (megabytes/second) at which data was read from all celldisks during the specified time range. Total Written Rate (Mb/S) The rate (megabytes/second) at which data was written on all celldisks during the specified time range. Alarms The number of warning, critical, and fatal alarms for the celldisks. When holding the cursor over one of the alarm counts, the dwell displays the most recent alarms invoked for these celldisks, sorted by severity. Clicking this field displays the Alarms list, which is listed by severity order. See the Foglight Online Help, “Monitoring System-Wide Alarms” topic for details on the alarm information. System Utilization CPU Load (%) The overall operating system CPU usage by all processes (including CPU usage by the database). Memory (%) The percentage, within the total memory, of memory consumed by all operating system processes (including the database). This value includes both RAM resident memory and swapped memory. Monitoring Status Indicator whether data is being collected. Selecting the Instance to be Monitored As the Databases dashboard displays by default all of the currently monitored databases, it is advisable to customize the dashboard’s view to display only the relevant instances, and then preview such instances to decide which to monitor. To select the instance to be monitored: 1 Display only instances relevant for your needs, using either of the following methods: • Filtering by severity — using the status indicators to display only instances of a specific database type that have a certain type of severity. For further details, see “Filtering the Display by Severity” on page 55. • Creating user-defined groups — using the Databases area to create groups that contain only the databases that need to be monitored for a specific need. For further details, see “Creating User-defined Database Groups” on page 55. 2 View the selected instances’ severity level, using the status indicators. 3 Click the requested instance to view a quick view panel with a preview of the instance’s most significant performance indicators. Managing Oracle Database Systems: User and Reference Guide 55 Chapter 2—Using Foglight for Oracle 4 Click Home Page on the quick view panel to access the Overview page of the selected instance or RAC. Filtering the Display by Severity The Status Indicators section can be used for filtering the Databases Group table to display only instances that meet a criterion set in this section. To filter the display by severity: 1 Click a a certain type of severity within a a specific database type (for example: Oracle instances whose severity level is Critical). The selected filter is shown in the table’s title. Selected filter 2 To select another filter, click the required status indicator (for example: Oracle > Normal severity). To display the entire list of databases, click Clear filters. Creating User-defined Database Groups Use the Databases area of the navigation pane to create, edit, and remove groups of database instances. Click Edit Group to add or remove database instances from the sub-group. To create a database group: 1 Select the parent database instance group, Database Group, under Databases in the navigation pane. Managing Oracle Database Systems: User and Reference Guide 56 Chapter 2—Using Foglight for Oracle 2 Click the button. The Add Group dialog box opens. 3 Type a name for the group in the Name field. 4 Type a description for the group (optional). 5 Select an instance in the Available column, and click [>] to move the instance to the Selected column. Alternatively, click [>>] to move all of the databases to the Selected column. 6 Click OK. The group name appears in the Database Groups list. The database instances are listed in the Databases pane. After adding one or more user-defined groups, it is possible to add sub-groups to these groups. To add sub-groups to a selected group: 1 Position the cursor on the requested group’s name. 2 Repeat step 2 to step 6. Note The Add Group dialog box displays the name of the parent group in the Parent field. Before proceeding, ensure that the sub-group is added to the requested group. To remove a database group: 1 Select the group to be removed. 2 Click the button. The Remove Groups dialog box appears, asking approval to remove the selected group with all its sub groups. 3 Click Remove. The group is removed from the Database Groups list. To edit a database sub-group: 1 Select the requested group. 2 Click the button. Managing Oracle Database Systems: User and Reference Guide 57 Chapter 2—Using Foglight for Oracle The Edit Group dialog box opens. 3 To add an instance to the group, select the requested instance in the Available column and click [>] to move the selected instance to the Selected column. 4 To remove an instance from the sub-group, select the requested instance in the Selected column, and click [<]. The selected instance moves to the Available column. 5 Click OK. Foglight for Oracle Overview Page The Foglight for Oracle Overview page provides various detailed views that are accessible using the toolbar. In addition, this page contains links to the Real-time and History home pages. The Overview page contents depends on whether a RAC (Real Application Cluster) or an Oracle database instance is selected. Breakdown and Baseline Chart Formats Several charts can display information in either of the following formats: • Baseline format — where a selected individual metric is displayed as a single line. • Breakdown format — a representation of actual activity of a metric or a set of metrics, compared with the typical behavior of these metrics for the selected time range. A breakdown display can also present a single metric divided into various components. For example, space utilization is broken down by the various components that occupy the space. Home Page Toolbar The Foglight for Oracle Overview page provides the toolbar buttons listed below. Name Description Summary Opens the Summary drilldown, which contains the following panels: • Overview page — see “Reviewing Overall Activity” on page 96. • History Summary — see “Reviewing Instance Historic Activity” on page 109. • Real-time Summary — see “Viewing the Foglight for Oracle Real-time Summary Page” on page 113. Performance Opens the Performance drilldown, which displays performance-related data at the following levels: • High-level data — available for all instances. • In-depth data, delivered by means of the Cluster view or Instance view tree — available only for instances or RACs that have Performance Analysis installed and configured. Managing Oracle Database Systems: User and Reference Guide 58 Chapter 2—Using Foglight for Oracle Name Description Activity Opens the Oracle Activity Drilldown, which contains the following panels: • Usability — see “Reviewing Instance Usability” on page 123 and “Reviewing Cluster Usability” on page 124. • Locks — see “Reviewing Locks Activity at the Cluster Level” on page 125 and “Reviewing Locks Activity at the Instance Level” on page 128. • Sessions — see “Viewing Detailed Sessions Data” on page 129. • Overall Datafiles I/O — see “Viewing Datafiles I/O Activity at the Cluster Level” on page 133 and “Viewing Datafiles I/O Activity at the Instance Level” on page 134. • User-defined Collections — see “Viewing User-Defined Collections” on page 134. Pluggable Databases Oracle instance 12C - pluggable only — Opens the Pluggable Databases Drilldown, which contains the following components: • Pluggable Databases List table — see “Pluggable Databases List table” on page 135. • Overview tab — see “Overview tab” on page 135. • Activity within Instance — shows the selected pluggable databases’ activity as part of the entire instance activity. For details, see “Activity within Instance/Cluster tab” on page 136. Storage Opens the Storage Drilldown, which contains the following panels: • Tablespaces and Datafiles — see “Tablespaces and Datafiles Panel” on page 137. • Redo Logs — see “Redo Logs Panel” on page 146. • Archive Logs — see “Archive Logs Panel” on page 148. • Invalid Objects — see “Invalid Objects Panel” on page 149. • Database Backup — see “Database Backup Panel” on page 150. The Storage panels are common for both RAC and all of the instances. Configuration Opens the Configuration drilldown, which contains the panels listed below. • SGA — see “Reviewing SGA Settings at the RAC Level” on page 151. • Initialization Parameters — see “Reviewing the Initialization Parameters” on page 153. Alerts Opens the Alerts drilldown, which contains the following panels: • Error Log — see “Reviewing the Error Log” on page 155. • Alert log — see “Reviewing the Alert Log” on page 156. Exadata In RACs that have Exadata configured, opens the Exadata drilldown, which contains the following panels: • Summary • Performance • InfiniBand Topology Managing Oracle Database Systems: User and Reference Guide 59 Chapter 2—Using Foglight for Oracle Reviewing Performance-related Data The Performance drilldown provides iterative access (up to three levels) to any of the key dimensions associated with Oracle database activity, based on the OLAP multidimensional model. Domain nodes offer a hierarchical view of all types of Oracle activity characteristics. Selecting a dimension from the Instance View tree determines what subset of activity Foglight for Oracle displays. Iterative investigations of domains of interest provide increasingly refined focus and diagnosis. This analysis can eventually result in the identification of the performance issue’s root cause, thereby enabling the resolution of the issue by means of the appropriate Quest tools. The root node to which the display opens is the entire instance activity level, to which the various dimensions aggregate. This level is the only one available if Performance Analysis is not installed and configured. For details, see “Investigating Performance Using the Node Tree” on page 59. Other menus, for which data is retrieved only with Performance Analysis installed and configured, display the various resource types and the wait events associated with each resource. Important If Performance Analysis is installed and configured, the Change Tracking view is displayed beside the Instance Overview, enabling you to keep track of all changes that took place within several, predefined categories. For details, see “Tracking Changes within Selected Categories” on page 61. Investigating Performance Using the Node Tree Expanding the tree shifts the application focus to that of the selected dimension that has been drilled-down to. The tree-pattern iterative drill-down is a two-node process, where the first node is the dimension type selection and the second is the dimension value. Important The default dimensions are different in ERP-enabled clusters. For details, see “Investigating the node tree in ERP-enabled clusters” on page 61. For example, to begin the installation by first identifying the most active DB User, follow the steps below: 1 Expand the instance view node to reveal the list of available dimensions. DB Users node Managing Oracle Database Systems: User and Reference Guide 60 Chapter 2—Using Foglight for Oracle 2 Click the DB Users node to display the most active database users; that is, the database users who consumed the highest amount of the selected resource. 3 Select the first user, to focus the entire window on this user's activity. 4 Identify the most resource-intensive SQL statement that this specific user has executed, by expanding the user node and then clicking the SQL statements dimension node. This will display the most active SQL statements executed by this user. 5 Click an SQL statement, to focus the entire window on that statement’s activity. 6 Click Client Machines under the selected SQL Statement, to view the computers on which the statement was run. In a similar manner, such iterative drilldowns can be carried out into any Oracle dimension of interest, to gain a complete understanding of the causes of its behavior. The following list details practical monitoring and administrative scenarios, where Performance Analysis dimensions provide the key insight for resolving the performance issue: • Finding the top resource-consuming SQL statements executed by a specific user • Finding the users that consume the highest amount of resources when executing a specific program, and (optionally) identifying the SQL statements causing the resource usage • Within the Oracle Application, finding the most active ORAAPPS users running a specific form The default Oracle dimensions are: • Instance (for RAC configurations) — SID identifying the cluster member instance • PL/SQL Blocks — the blocks that comprise the PL/SQL code used, grouped by: • Procedure • Function • Package • Trigger • Anonymous block • SQL Statements — the executed SQL queries Managing Oracle Database Systems: User and Reference Guide 61 Chapter 2—Using Foglight for Oracle • Programs — names of the programs that connected to Oracle and executed the SQL statements Note If Oracle fails to report a valid program name, and the program is running on the database host, the program will be named according to the running program executable. • OS Users — operating system users running the client program • DB Users — Oracle users used by the program to connect to the Oracle instance • Client Machines — the computers where the client executable (connected to Oracle) is running • Actions — the name of the currently executing action, as set by calling the DBMS_ APPLICATION_INFO.SET_ACTION procedure • Modules — the name of the currently executing module as set by calling the DBMS_ APPLICATION_INFO.SET_MODULE procedure • Command Types — executed SQL command type; for example INSERT and SELECT • Consumer Groups — displays a collection of users with similar requirements for resource consumption • Services — displays Oracle's internal services: • SYS$BACKGROUND — used by the background processes only • SYS$USERS — the default service for users that connect locally • User-defined services — enables the identification of sessions that connect to such services Note When data received from Oracle contains empty values (for example, Oracle background processes that have no specific database user), the designation appears in the dimension tree of the history dimension. Drilling down within this domain reveals its identifiable components. Investigating the node tree in ERP-enabled clusters Foglight for Oracle Performance Analysis incorporates Enterprise Resource Planning (ERP) advanced analytic features. These features can be used for associating database activity, generated by the ERP application, with the corresponding ERP context (for example, ERP Users, Forms and Applications), as well as advanced ERP performance advisories. ERP support is available for PeopleSoft version 8.0 or later and Oracle Applications version 11.0.x, 11i or 12. Tracking Changes within Selected Categories The Change Tracking view is displayed and available only when Performance Analysis is installed and configured. Managing Oracle Database Systems: User and Reference Guide 62 Chapter 2—Using Foglight for Oracle Categories section Use this view to display the change history for one or more of the following categories, which are predefined in Performance Analysis: • Execution plan — SQL statements whose execution plans have changed. Unsuitable execution plans can result in SQL performance degradation. • Oracle configuration — instance-level configuration, such as init.ora parameters, location of files (data, log, and control), size and status, and rollback segments. • Oracle schema — any schema object, such as tables and indexes and their partitions, clusters, constraints, views, and materialized views. • System configuration — hardware and operating system configuration, such as OS global parameters, amount of RAM, and number of CPU units. • User-defined — used for documenting changes inserted by the user, which can affect performance; for example, installing an operating system patch. Recording this activity can assist in determining whether a change in performance can be directly associated with this event. Important The Change Tracking tool is an integrated monitoring feature within Performance Analysis, which is available only if the Performance Analysis repository is installed. Note The data displayed in the Change Tracking view is independent of the resources selected in the Performance drilldown (Network, DB Buffer, and so on), and is relevant only to the categories selected for display in the Categories section. Investigating Performance Using the Various Resource Types Use the Performance toolbar to access a panel that displays the wait events associated with the specific resource that bears the command name, such as CPU, Lock, and Latch. Managing Oracle Database Systems: User and Reference Guide 63 Chapter 2—Using Foglight for Oracle Reviewing the Overall Workload Use the Workload panel to review the overall workload and wait events. This panel contains the following sections: • Average Active Sessions chart — see “The workload-related resource consumption (Average Active Sessions) chart” on page 63. • Lower section — its contents depend on where the cursor is positioned, as described in the following sections: • “Positioning the cursor over the instance or cluster view (root)” on page 67 • “Positioning the cursor over a specific dimension” on page 67 • “Metrics displayed in the CPU > Overview pane” on page 67 The workload-related resource consumption (Average Active Sessions) chart The Average Active Sessions chart displays the metrics listed below, in either of the following view: • Baseline — displays the entire workload. Important The baseline view is displayed only when the cursor is positioned over the Instance View (root). • Breakdown — where individual sub-categories of wait category values are displayed as part of the total activity. Positioning the cursor over the instance or cluster view (root) When the cursor is positioned over the instance or cluster view (root), the Overview and Highlights section is displayed, showing the following panes: • Overview — displays the metric selected in the metric table. Managing Oracle Database Systems: User and Reference Guide 64 Chapter 2—Using Foglight for Oracle • Highlights — displays situations where an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload. This is likely to be exceptional and significant behavior. Positioning the cursor over a dimension type Upon positioning the cursor over a dimension type (for example, OS Users or DB Users), the lower section displays the following panes: • Top dimension section — includes a chart and a metrics table that display information about the top-consuming entities within the dimension selected on the left (for example, Top DB Users), the lower section contains the following panes: • Chart — displays the CPU wait for the dimension in the selected row. • Table for top resources — for details, see “Table for top resources (Workload panel)” on page 64. Positioning the cursor over a specific dimension When the cursor is positioned over a specific dimension (for example, a specific SQL statement or DB user), the lower section contains the following panes: • Overview — displays the metric selected in the metric table. • Workload-related Metrics table — displays the CPU-related wait events. For details, see “Metrics displayed in the Workload > Overview pane” on page 64. The contents of this table can be customized by clicking Select metric ( adding or removing metrics using the Metrics Selection dialog box. ) and Metrics displayed in the Workload > Overview pane The Workload > Overview pane includes a chart that displays the metrics selected in the CPUrelated Metrics table. The chart contains the following information: • Metric — the metric name; can be one of the following metrics: • SQL*Net Roundtrips to/from Client — total number of SQL*Net messages sent to and received from the client. • CPU Usage — see “CPU Usage” on page 209. • Physical Reads — see “Physical Reads” on page 218. Table for top resources (Workload panel) The table displayed for top resources includes the following columns: • Actions — the actions that can be carried out for the selected resource within the selected row Note The actions that can take place when positioning the cursor on the top resource dimension and clicking the cogwheel button ( ) depend on whether Performance Analysis is installed and configured. For details, see “The Actions row” on page 65. • Resource type name — the name of the selected resource type; for example, database name or OS user name • Source — the metric’s source (OS, Oracle and so on) • Total — the metric’s total value Managing Oracle Database Systems: User and Reference Guide 65 Chapter 2—Using Foglight for Oracle The Actions row When positioning the cursor on the top resource dimension (the root of the SQL Statements, Programs and so on), and clicking the cogwheel button ( ), a menu appears with several options. The options available in this menu vary depending on whether Performance Analysis is installed and configured, as detailed in the following sections: • “Actions available with Performance Analysis installed and configured” on page 65 • “Actions available without Performance Analysis” on page 65 Actions available with Performance Analysis installed and configured If Performance Analysis is installed and configured, the Actions available upon clicking the cogwheel button are as follows: • Tune SQL — clicking this option downloads the foglight.questsqloptimizer_ora file. If SQL Optimizer for Oracle is installed, this file can be used for optimizing the SQL query. • Roll Up — if the resource was reached by drilling down several levels, using this option displays the resource tree node with the selected resource at the top of the node. • Execution Plan — provides access to the Execution Plan history feature. Note Note Using the Execution Plan history feature requires installing a Performance Repository for the selected instance. The Tune SQL and Execution Plan options are available only when the cursor is positioned on a top resource view. Actions available without Performance Analysis If Performance Analysis is not installed and configured, only the SQL Statements dimension is available. Therefore, the actions available upon clicking the cogwheel button are as follows: • View Full Text — displays the SQL query’s full text • Roll Up Reviewing CPU Usage Use the CPU panel to review the CPU usage and CPU wait events. Managing Oracle Database Systems: User and Reference Guide 66 Chapter 2—Using Foglight for Oracle Important The CPU panel is only available if Performance Analysis is installed and configured. This panel contains the following sections: • Average Active Sessions chart — see “The CPU-related resource consumption (Average Active Sessions) chart” on page 66. • Lower section — its contents depend on where the cursor is positioned, as described in the following sections: • Positioning the cursor over the instance or cluster view (root) • Positioning the cursor over a specific dimension • Metrics displayed in the CPU > Overview pane The CPU-related resource consumption (Average Active Sessions) chart The Average Active Sessions chart displays the metrics listed below, in either of the following view: • Baseline — displays only the CPU Usage metric. Important The baseline view is displayed only when the cursor is positioned over the Instance View (root). • Breakdown — where individual sub-categories of CPU wait category values are displayed as part of the total activity. The following metrics are displayed in the Average Active Sessions chart: Managing Oracle Database Systems: User and Reference Guide 67 Chapter 2—Using Foglight for Oracle • CPU Usage — see “CPU Usage” on page 209. • CPU Wait — see “CPU Wait” on page 209. Note The CPU Wait value should be inspected in conjunction with the value of the Run Queue Length metric (available only when Foglight Performance Analysis is installed and configured). • Non-CPU activities — the time spent on active wait events, excluding CPU wait events. This metric shows the ratio between time spent only on CPU-related wait events and the total time spent on wait events. Positioning the cursor over the instance or cluster view (root) When the cursor is positioned over the instance or cluster view (root), the Overview and Highlights section is displayed, showing the following panes: • Overview — displays the metric selected in the metric table. • Highlights — displays situations where an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload. This is likely to be exceptional and significant behavior. • CPU-related Metrics table — displays the CPU-related wait events. For details, see “Metrics displayed in the CPU > Overview pane” on page 67. The contents of this table can be customized by clicking the Select metric button and adding or removing the metrics using the Metrics Selection dialog box. Positioning the cursor over a dimension type Upon positioning the cursor over a dimension type (for example, OS Users or DB Users), the lower section displays the following panes: • Top dimension section — includes a chart and a metrics table that display information about the top-consuming entities within the dimension selected on the left (for example, Top DB Users), the lower section contains the following panes: • Chart — displays the CPU wait for the dimension in the selected row. • Table for top resources — for details, see “Table for top resources (CPU panel)” on page 68. Positioning the cursor over a specific dimension When the cursor is positioned over a specific dimension (for example, a specific SQL statement or DB user), the lower section contains the following panes: • Overview — displays the metric selected in the metric table. • CPU-related Metrics table — displays the CPU-related wait events. For details, see “Metrics displayed in the CPU > Overview pane” on page 67. The contents of this table can be customized by clicking the Select metric button and adding or removing metrics using the Metrics Selection dialog box. Metrics displayed in the CPU > Overview pane The CPU > Overview pane includes a chart that displays the metrics selected in the CPU-related Metrics table. The chart contains the following information: Managing Oracle Database Systems: User and Reference Guide 68 Chapter 2—Using Foglight for Oracle • Metric — the metric name; can be one of the following metrics: • Row Source Ratio — see “Row Source Ratio” on page 218 • Degree of Parallelism — the total number of server processes (master and slaves) assigned to handle a single operation. Table for top resources (CPU panel) The table displayed for top resources includes the following columns: • Actions — the actions that can be carried out for the selected resource within the selected row Note The actions that can take place when positioning the cursor on the top resource dimension and clicking the cogwheel button ( ) depend on whether Performance Analysis is installed and configured. For details, see “The Actions row” on page 65. • Resource type name — the name of the selected resource type; for example, database name or OS user name • CPU Wait — see “CPU Usage” on page 209 • I/O Wait — see “IO Wait” on page 212 • User Calls per Parse — see “User Calls per Parse” on page 221 • Calls — see “Calls” on page 208 • Direct Path I/O — see “Direct Path IO Wait” on page 210 Reviewing Disk I/O Activity Use the Disk I/O panel to view I/O related data such as wait events, load balance, and physical reads and writes. The content of this panel depends on whether a RAC (Real Application Cluster) or a database instance is selected. For details about the instance view of this panel, see “Disk I/O Panel in instance mode” on page 74. Managing Oracle Database Systems: User and Reference Guide 69 Chapter 2—Using Foglight for Oracle Disk I/O panel in Whole Cluster mode In Whole Cluster mode, the Disk I/O panel displays information aggregated from the individual instances to the cluster level. The bottom balance chart displays the functionality of the individual instances according to the selected balance criterion. The contents of this panel vary between instances monitored using Performance Analysis and instances that do not have Performance Analysis installed and configured, as described in the following sections: • “I/O panel without Performance Analysis” on page 69 • “I/O panel with Performance Analysis installed and configured” on page 70 I/O panel without Performance Analysis If Performance Analysis is not installed and configured, the Disk I/O panel contains data only when the cursor is positioned over the Instance View (root), as follows: • Disk I/O-related resource consumption chart (Average Active Sessions) — displays the metrics listed below, in either of the following manners: • A baseline value — where individual sub-categories of an I/O wait category are selected • A breakdown — where individual sub-categories of I/O wait category values are displayed as part of the total activity The metrics displayed in the Average Active Sessions chart are: • IO-related metrics, whose value is greater than zero. For details, see “Metrics displayed in the I/O > Overview pane” on page 71. Managing Oracle Database Systems: User and Reference Guide 70 Chapter 2—Using Foglight for Oracle • Non IO Wait — total active time spent waiting for all resources, excluding IO. • Overview — displays the metric selected in the metric table • I/O-related Metrics table — displays the I/O-related wait events. The contents of this table can be customized by clicking the Select metric button and adding or removing the requested metrics using the Metrics Selection dialog box. I/O panel with Performance Analysis installed and configured The I/O panel contains the following sections: • I/O-related resource consumption chart (Average Active Sessions) — see “I/O-related resource consumption chart (Average Active Sessions)” on page 70 • Lower section — the contents of the lower section depends on where the cursor is positioned, as described in the following sections: • Positioning the cursor over the instance or cluster view (root) • Positioning the cursor over a specific dimension • Metrics displayed in the CPU > Overview pane I/O-related resource consumption chart (Average Active Sessions) The Average Active Sessions chart displays the metrics listed below, in either of the following views: • Baseline — where individual sub-categories of an I/O wait category are selected • Breakdown — where individual sub-categories of I/O wait category values are displayed as part of the total activity The metrics displayed in the Average Active Sessions chart are: • IO-related metrics, whose value is greater than zero. For details, see “Metrics displayed in the I/O > Overview pane” on page 71. • Non IO Wait — total active time spent waiting for all resources, excluding IO Positioning the cursor over the cluster view (root) When the cursor is positioned over the instance view (root), the Overview and Highlights section is displayed, showing the following panes: • Overview — contains the following components: • The I/O-related Metric table — displays the I/O-related wait events. For details, see “Metrics displayed in the I/O > Overview pane” on page 71. The contents of this table can be customized by clicking the Select metric button and adding or removing the requested metrics using the Metrics Selection dialog box • A chart — displays the metric selected in the metric table • I/O Metrics — contains the following sections: • Physical Reads • Physical Writes • Balance — displays information for each database instance, and comprises the following components: Managing Oracle Database Systems: User and Reference Guide 71 Chapter 2—Using Foglight for Oracle Balance Criterion chart — the balance criterion can be changed by clicking the metric name and selecting a different criterion, as described in the table below. Cluster Balance (%) — the average load balance value for all instances for the selected criterion. To drill down to details on any instance, click an instance bar to view the selected node in a single instance/RAC node view. The metrics plotted in the Balance Criterion chart are the same as described in the Metrics table, except for the Sort Overflow Ratio metric. • Highlights — displays situations where an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior Positioning the cursor over a dimension type Upon positioning the cursor over a dimension type (for example: OS Users or DB Users), the lower section displays the following panes: • Top dimension section — includes a chart and a metrics table for the top-consuming entities within the dimension selected on the left (for example, Top DB Users). The lower section contains the following panes: • Chart — displays the IO-related wait for the dimension in the selected row • Table for top resources — see “Table for top resources (Disk I/O panel)” on page 72 Positioning the cursor over a specific dimension When the cursor is positioned over a specific dimension (for example: a specific SQL statement or DB user), the lower section contains the following panes: • Overview — displays the metric selected in the metric table • I/O-related Metrics table — displays the I/O-related wait events. For details, see “Metrics displayed in the I/O > Overview pane” on page 71. The contents of this table can be customized by clicking the Select metric button and adding or removing the requested metrics using the Metrics Selection dialog box. Metrics displayed in the I/O > Overview pane The I/O > Overview pane includes a chart that displays the metric selected in the Metrics table below, which contains by default the following metrics, in cluster view: • Average Read Time • Average Write Time • Block Reads • Cache Hit Ratio • Chained Row Ratio • Disk Utilization • Oracle Data I/O Operations • Oracle Physical I/O Operations Managing Oracle Database Systems: User and Reference Guide 72 Chapter 2—Using Foglight for Oracle • Paging • Physical Reads • Physical Writes • Sort Overflow Ratio Table for top resources (Disk I/O panel) The table displayed for top resources includes the following columns: • Actions — the actions that can be carried out for the selected resource within the selected row Note The actions that can take place when positioning the cursor on the top resource dimension and clicking the cogwheel button ( ) depend on whether Performance Analysis is installed and configured. For details, see “The Actions row” on page 65. • Resource type name — the name of the selected resource type; for example, database name or OS user name • IO Wait — time spent waiting for disk input/output operations to complete • SQL Executions — the number of SQL statements executed during the specified time range • Physical IO The Oracle Database RAC I/O contains the following sections: • Workload • Statistics and Highlights — including the following panes: • Overview • I/O Metrics • Highlights — displays situations in which an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior Managing Oracle Database Systems: User and Reference Guide 73 Chapter 2—Using Foglight for Oracle Workload Chart Pane The Total I/O Wait (Average Active Sessions) chart displays the metrics listed in the table below, in either of the following manners: • A baseline value — where individual sub-categories of an I/O wait category are selected • A breakdown — where individual sub-categories of I/O wait category values are displayed as part of the total activity The metrics displayed in the I/O Activity Chart are: • Total IO Wait • Datafile Write Wait • Multiblock Read Wait • Direct Path IO Wait • Control File IO Wait • DB File Parallel Read Wait • External File IO Wait • Miscellaneous IO Wait • Non IO Wait Managing Oracle Database Systems: User and Reference Guide 74 Chapter 2—Using Foglight for Oracle Overview pane The Overview pane includes a chart that displays the metric selected in the Metrics table. This table contains the following metrics: • Average Read Time • Average Write Time • Cache Hit Ratio • Chained Row Ratio • Physical Reads • Physical Writes • Sort Overflow Ratio Disk I/O Panel in instance mode In instance mode, the Disk I/O drilldown provides information using the following sections: • Average Active Sessions chart The content of this chart is identical to that of the Average Active Sessions chart in Whole Cluster mode. For more information, see “Workload Chart Pane” on page 73. • Resource Breakdown — a pie chart displaying the percentage of active time consumed by all resources. • Overview — includes the panes described in the following sections: • • • • • • Overview pane I/O Statistics pane Logical I/O pane Physical I/O pane Operations Pane Highlights — displays situations in which an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior Managing Oracle Database Systems: User and Reference Guide 75 Chapter 2—Using Foglight for Oracle Overview pane The Overview pane includes a metric table along with a chart, which displays the metric selected in the table, plotted over the specified time range. The Metrics table includes all metrics displayed on the RAC I/O Metrics pane, in addition to the following metric: • Sort Overflow Ratio — measures the portion of sorts that could not fit into a session's sort area and therefore used temporary segments. I/O Statistics pane The I/O Statistics pane provides several sections of information. Each section displays a bar representation of the information that is displayed graphically in its associated pane (for example, Logical I/O), with a display of real-time behavior, color-coded for severity. Click a metric value or bar on each of these sections to drill down to the associated pane. • Under the Logical I/O section: • Logical Reads • Block Changes • Under the Physical I/O section: • Physical Reads • Physical Writes • Redo Writes Managing Oracle Database Systems: User and Reference Guide 76 Chapter 2—Using Foglight for Oracle • Under the Performance section: • Average Read Time • Average Write Time • Cache Hit Ratio Logical I/O pane The Logical I/O pane displays the metrics either as a summary for the selected time range, or plotted over time. The Logical I/O pane provides information in the sections listed below. • Logical Reads — contains the following components: • Logical reads chart — can be displayed as a baseline value where individual metrics are selected, or as a breakdown where individual metric values are displayed as part of the total activity. For details, see “Logical Reads Chart” on page 76. • Logical reads table — for details, see “Logical Reads Table” on page 76. • Block Changes — contains the following components: • Block changes chart — can be displayed as a baseline value where individual metrics are selected, or as a breakdown where individual metric values are displayed as part of the total activity. For details, see “Block Changes Chart” on page 76. • Block changes table — for details, see “Block Changes Table” on page 77. • Read Consistency Overhead — the overhead resulting from running queries that process parts of the update and delete operations, in order to obtain consistent data. Logical Reads Chart • Session Logical Reads • DB Block Gets • Consistent Gets Logical Reads Table Name Description Metric Name The name of the metric: • Session Logical Reads • DB Block Gets • Consistent Gets Value The total number of reads or gets for the specified timeline. Baseline The baseline value for the metric. Alarms The number of warning, critical, and fatal alarms for the selected tablespace. Block Changes Chart • Total Changes — the summary of database block changes and consistent changes. Managing Oracle Database Systems: User and Reference Guide 77 Chapter 2—Using Foglight for Oracle • DB Block Changes • Consistent Changes Block Changes Table Name Description Metric Name The name of the metric: • DB Block Changes • Consistent Changes Value The total number of reads or gets for the specified timeline. Baseline The baseline value for the metric. Alarms The number of warning, critical, and fatal alarms for the selected tablespace. Physical I/O pane The Physical I/O pane displays the metrics in either of the following forms: summary for the selected time range, and plotted over time. The Physical I/O pane provides information by means of the following sections: • Physical Reads — contains the following components: • Physical reads chart — can be displayed as a baseline value where individual metrics are selected, or as a breakdown where individual metrics values are displayed as part of the total activity. For details, see “Physical Reads Chart” on page 77. • Physical reads table — for details, see “Physical Reads Table” on page 78. • Physical Writes — contains the following components: • Physical writes chart — can be displayed as a baseline value where individual metrics are selected, or as a breakdown where individual metrics values are displayed as part of the total activity. For details, see “Physical Writes Chart” on page 78. • Physical writes table — for details, see “Physical Writes Table” on page 78. Physical Reads Chart Name Description Total Physical Reads The total number of data blocks that were read using database physical reads. Physical Reads Direct The total number of data blocks read directly from disk into the PGA, bypassing the buffer cache. Physical Reads Cache Total number of data blocks read from disk into the buffer cache. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Physical Reads Table Name Description Metric Name The name of the metric: • Physical Reads Direct • Physical Reads Cache • Average Read Time Value The total number of reads or gets for the specified timeline. Baseline The baseline value for the metric. Alarms The number of warning, critical, and fatal alarms for the selected tablespace. Physical Writes Chart Name Description Total Physical Writes The summary of database physical writes to the disk, whether directly or from the buffer cache. Physical Writes Direct The number of data blocks written directly to disk, bypassing the buffer cache. Physical Writes Cache Total number of data blocks written to disk from the buffer cache. Physical Writes Table Name Description Metric Name The name of the metric: • Physical Writes Direct • Physical Writes Cache • Average Write Time Value The total number of reads or gets for the specified timeline. Baseline The baseline value for the metric. Alarms The number of warning, critical, and fatal alarms for the selected tablespace. Operations Pane The Operations pane displays the individual I/O operations, such as full table scan. The Operations pane provides information by means of the following sections: • Table Fetches — contains the following components: • Table Fetches chart — for details, see “Table Fetches Chart” on page 79. 78 Managing Oracle Database Systems: User and Reference Guide 79 Chapter 2—Using Foglight for Oracle • Table Fetches table — for details, see “Table Fetches Table” on page 79. • Disk Sorts — contains the following components: • Disk Sorts chart — can be displayed as a baseline value where individual metrics are selected, or as a breakdown where individual metrics values are displayed as part of the total activity. For details, see “Disk Sorts Chart” on page 79. • Disk Sorts table — for details, see “Disk Sorts Table” on page 79. Table Fetches Chart • Table scan rows gotten • Table fetch by row ID • Table fetch continued row Table Fetches Table Name Description Metric Name The name of the metric: • Table scan rows gotten • Table scan blocks gotten • Table fetch by row ID • Table fetch continued row • Chained row ratio • Full scan ratio Value The total number of reads or gets for the specified timeline. Baseline The baseline value for the metric. Alarms The number of warning, critical, and fatal alarms for the selected tablespace. Disk Sorts Chart • Total Sorts • Disk Sorts • Memory Sorts Disk Sorts Table Name Description Metric Name The name of the metric: • Disk Sorts • Memory Sorts • Disk Sorts Ratio — the ratio of disk sorts to memory sorts. Value The total number of reads or gets for the specified timeline. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Name Description Baseline The baseline value for the metric. Alarms The number of warning, critical, and fatal alarms for the selected tablespace. Network panel The Network panel displays the network-related wait events. This panel includes the following sections: • Workload (Average Active Sessions) chart — see “The Workload chart” on page 80 • Overview and Highlights section — includes the following panes: • Overview — displays the metric selected in the metric table • Highlights — displays situations where an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior • Metric table — see “The Metrics table” on page 81 The Workload chart The Workload (Average Active Sessions) chart displays the metrics listed in the table below, in either of the following manners: • A baseline value — displays only the Network Wait metric • A breakdown — where individual sub-categories of network wait category values are displayed as part of the total activity 80 Managing Oracle Database Systems: User and Reference Guide 81 Chapter 2—Using Foglight for Oracle The Workload (Average Active Sessions) chart displays metrics such as: • Network Wait • SQL *Net • DB Link Wait Positioning the cursor over a dimension type Upon positioning the cursor over a dimension type (for example: OS Users or DB Users), the lower section displays the following panes: • Top dimension section — includes a chart and a metrics table for the top-consuming entities within the dimension selected on the left (for example, Top DB Users). The lower section contains the following panes: • Chart — displays the network-related wait for the dimension in the selected row • Table for top resources — see “Table for top resources (Network panel)” on page 81 The Metrics table Use the Metrics table to select one of the following metrics: • Network Wait — time spent waiting for messages to be sent or received over the network interface. • Packets — total number of packets transmitted and received by the operating system on the various network interfaces. • SQL*Net Roundtrips to/from Client — total number of SQL*Net messages sent to and received from the client. • User Calls — number of user calls such as login, parse, fetch, or execute. This metric value represents the rate at which the user application addresses Oracle. Table for top resources (Network panel) The table displayed for top resources includes the following columns: • Actions — the actions that can be carried out for the selected resource within the selected row Note The actions that can take place when positioning the cursor on the top resource dimension and clicking the cogwheel button ( ) depend on whether Performance Analysis is installed and configured. For details, see “The Actions row” on page 65. • Resource type name — the name of the selected resource type; for example, database name or OS user name Note The columns detailed below are displayed only if Performance Analysis is installed and configured. • SQL Executions — the number of SQL statements executed during the specified time range • SQL *Net Roundtrips to/from Client — the total number of Net8 messages that were sent to and received from the client. • User Calls — number of user calls such as login, parse, fetch, or execute. This metric value represents the rate at which the user application addresses Oracle. Managing Oracle Database Systems: User and Reference Guide 82 Chapter 2—Using Foglight for Oracle • Degree of Parallelism — the total number of server processes (master and slaves) assigned to handle a single operation. • Kbytes Sent via SQL *.Net to Client — the total traffic, in kilobytes, sent to the Client using SQL *.Net. DB Buffer panel The DB Buffer panel displays the database buffer-related wait events. This panel includes the following sections: • Workload (Average Active Sessions) chart — see “The Workload chart” on page 82 • Overview and Highlights section — includes the following panes: • Overview — displays the metric selected in the metric table • Highlights — displays situations in which an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior • Metric table — see “The Metrics table” on page 83 The Workload chart The Workload (Average Active Sessions) chart displays the metrics listed in the table below, in either of the following manners: • A baseline value — displays only the Buffer Wait metric Managing Oracle Database Systems: User and Reference Guide 83 Chapter 2—Using Foglight for Oracle • A breakdown — where individual sub-categories of network wait category values are displayed as part of the total activity The Workload (Average Active Sessions) chart displays metrics such as: • Buffer Wait (appears only in Baseline view) • Non-buffer Activities • Other DB Buffer Wait • Free Buffer Wait • Buffer Busy Wait The Metrics table Use the Metrics table to select one of the following metrics: • Cache Hit Ratio • Sort Overflow Ratio Lock panel The Lock panel displays the database’s lock-related wait events. This panel includes the following sections: • Locks Activity — contains the Lock Wait (Average Active Sessions) chart; see “The Lock Wait chart” on page 84 • Statistics and Highlights section — includes the following panes: • Overview — displays the metric selected in the metric table Managing Oracle Database Systems: User and Reference Guide 84 Chapter 2—Using Foglight for Oracle • Lock Metrics — the contents of this section varies between RAC and instance views. For details, see “Lock Metrics in Whole Cluster mode” on page 84 and “Lock metrics in Instance mode” on page 84 • Metric table — see “The Metrics table” on page 83 The Lock Wait chart The Lock Wait (Average Active Sessions) chart displays the metrics listed in the table below, in either of the following manners: • A baseline value — displays only the Lock Wait metric • A breakdown — displays the lock wait value as part of the total activity, by comparing it with the non-lock wait The Lock Wait (Average Active Sessions) chart displays metrics such as: • Lock Wait • Non Lock Wait Lock Metrics in Whole Cluster mode The Lock Metrics pane contains the following sections: • Enqueue Waits chart — displays the number of lock requests that escalated to a blocking lock, plotted over time. • Balance — the balance criterion is selected from the values listed below, and is displayed as a vertical bar for each Oracle node within the RAC. • Waits for Locks — number of wait events per lock • Enqueue Waits — number of lock requests that escalated to a blocking lock • Lock Efficiency — percentage of lock requests that did not escalate to a blocking lock The Cluster Balance (%) is displayed as a number and bar gauge value. Lock metrics in Instance mode The Lock Metrics pane contains the following sections: • Enqueue Waits chart — displays the number of lock requests that escalated to a blocking lock, plotted over time. • Blocked Segments — a table listing the name of each segment, with the number of times a wait event occurred during this time period. The Blocked Segments table is sorted according to the Wait column. • Locks Efficiency — the percentage of lock requests that did not escalate to a blocking lock. This bar displays real-time behavior, color-coded for severity. • Lock Wait Time Breakdown — a pie chart showing the percentage of waits for: • • • • System TM — DML enqueue TX — Transaction UL — User-defined Lock Managing Oracle Database Systems: User and Reference Guide 85 Chapter 2—Using Foglight for Oracle Positioning the cursor over a dimension type Upon positioning the cursor over a dimension type (for example: OS Users or DB Users), the lower section displays the following panes: • Top dimension section — includes a chart and a metrics table for the top-consuming entities within the dimension selected on the left (for example, Top DB Users). The lower section contains the following panes: • Chart — displays the lock-related wait for the dimension in the selected row • Table for top resources — see “Table for top resources (Lock panel)” on page 85 The Metrics table In the Lock panel, the Metrics table contains a single metric: Average Lock Duration. This metric indicates the average time spent (ms) on a resource that was locked (held in contention). Lock waits occur when a session attempts to acquire a lock that is already held by another session. One session will be blocked until the blocking session (the first to acquire a lock on the resource) releases the lock. High values of lock duration can indicate aggressive lock incidence, which may be improved upon tuning. Table for top resources (Lock panel) The table displayed for top resources includes the following columns: • Actions — the actions that can be carried out for the selected resource within the selected row Note The actions that can take place when positioning the cursor on the top resource dimension and clicking the cogwheel button ( ) depend on whether Performance Analysis is installed and configured. For details, see “The Actions row” on page 65. • Resource type name — the name of the selected resource type; for example, database name or OS user name Note The columns detailed below are displayed only if Performance Analysis is installed and configured. • Lock Wait — the lock-related wait events. • SQL Executions — the number of SQL statements executed during the specified time range. • Average Lock Duration — the average length of time for a lock, in milliseconds. • Enqueue Requests — the total number of table or row locks acquired. • Enqueue Waits — the total number of times an enqueue request or conversion resulted in a wait event. • Degree of Parallelism — the total number of server processes (master and slaves) assigned to handle a single operation. Latch panel The Latch panel displays the database’s latch-related wait events. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle This panel includes the following sections: • Workload — contains the Workload (Average Active Sessions) chart; see “The Workload chart” on page 86 • Statistics and Highlights section — includes the following panes: • Overview — displays the metric selected in the metric table, that is, Latch Wait • Highlights — displays situations in which an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior • Metric table — displays the Latch Wait metric The Workload chart The Workload (Average Active Sessions) chart displays the metrics listed in the table below, in either of the following manners: • A baseline value — displays only the Latch Wait metric • A breakdown — displays the latch wait value as part of the total activity, by comparing it with the non-latch wait The Workload (Average Active Sessions) chart displays metrics such as: • Latch Wait • Non Latch Wait 86 Managing Oracle Database Systems: User and Reference Guide 87 Chapter 2—Using Foglight for Oracle Positioning the cursor over a dimension type Upon positioning the cursor over a dimension type (for example: OS Users or DB Users), the lower section displays the following panes: • Top dimension section — includes a chart and a metrics table for the top-consuming entities within the dimension selected on the left (for example, Top DB Users). The lower section contains the following panes: • Chart — displays the lock-related wait for the dimension in the selected row • Table for top resources — see “Table for top resources (Latch panel)” on page 87 The Metrics table In the Latch panel, the Metrics table contains a single metric: Latch Wait. This metric indicates the time spent by the session being blocked by a latch, waiting for it to be released. Latches need not be locked for the duration of a transaction. They are low-overhead, short-term memory synchronization objects. They are used mostly to protect a row when it is queried for a connection. Table for top resources (Latch panel) The table displayed for top resources includes the following columns: • Actions — the actions that can be carried out for the selected resource within the selected row Note The actions that can take place when positioning the cursor on the top resource dimension and clicking the cogwheel button ( ) depend on whether Performance Analysis is installed and configured. For details, see “The Actions row” on page 65. • Hash Value Note The columns below are displayed only if Performance Analysis is installed and configured. • SQL Statement • Latch Wait — the time spent by the session being blocked by a latch • SQL Executions — the number of SQL statements executed during the specified time range. • Degree of Parallelism — the total number of server processes (master and slaves) assigned to handle a single operation. Redo panel The contents of the Redo panel depends on whether a RAC (Real Application Cluster) or an individual instance is selected, as described in the following sections: • Redo panel at the RAC level • Redo panel at the instance level — for details, see “Redo Panel at the Instance Level” on page 90 Managing Oracle Database Systems: User and Reference Guide 88 Chapter 2—Using Foglight for Oracle Redo panel at the RAC level In the Whole Cluster mode, the Redo panel comprises the following sections: • Redo and Archive Activity chart — for details, see “Redo and Archive Activity Chart” on page 89. • Statistics and highlights section — contains the following panes: • Overview — contains the Metrics table and a chart that provides a representation of the metric selected in the table. For details, see “Metrics table” on page 89. • Redo Metrics — for details, see “Redo Metrics pane” on page 89. • Highlights — displays situations where an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior. Managing Oracle Database Systems: User and Reference Guide 89 Chapter 2—Using Foglight for Oracle [ Redo and Archive Activity Chart The Redo and Archive Activity chart can be displayed as a baseline value where individual metrics are selected, or as a breakdown where individual metrics values are displayed as part of the total activity. Redo and Archive Activity • Total Redo Wait (displayed only in the baseline view) • Archive Wait • Log Buffer Wait • Log Switch Wait • Log Write Wait • Non Redo Wait (displayed only in the Breakdown view, for showing the ratio between time spent on redo wait events and the total time spent on wait events) Metrics table The Metrics table includes the following metrics: • Average Redo Write Time • Log Switch Wait Ratio • Redo Wait • Redo Writes Redo Metrics pane The Redo Metrics pane includes the following sections: • Redo Writes chart — the total number of write operations issued by LGWR (Log Writer) to the Redo log files. This value is Plotted over time. Managing Oracle Database Systems: User and Reference Guide 90 Chapter 2—Using Foglight for Oracle • Balance section — comprises the following components: • Balance Chart • Cluster Balance (%) — the average load balance value for all instances for the selected criterion. To drill down to details on any instance, click the specific instance bar. The Redo panel at the instance level appears. For details, see “Redo Panel at the Instance Level” on page 90. Balance Chart The displayed balance criterion can be changed by clicking the metric name and selecting a different criterion. • Average Redo Write Time • Redo Waits • Redo Writes • Archive Wait Redo Panel at the Instance Level The Redo panel at the instance level comprises the following sections: • Resource Consumption — contains the Average Active Sessions chart, which can be displayed as a baseline value where individual sub-categories of a Redo wait category are selected, or as a breakdown where individual sub-categories of a Redo wait category values are displayed as part of the total activity. The redo and archive activity chart also appears in the Redo and Archive Performance for a RAC panel. For more information, see “Redo and Archive Activity Chart” on page 89. • Overview section — contains the following panes: • Overview — contains the Metrics table and a chart that provides a representation of the metric selected in the table. The Metrics table includes the same metrics as that of the Metrics table for the RAC-level panel. For details, see “Metrics table” on page 89. • The Redo Writes pane — contains the following charts: Redo Log Write Average Redo Write Time • The Log Switches pane — contains the following charts: Number of Switches — the number of log switches, plotted over time. Log Switches Latency — the amount of time (in milliseconds) spent on log switches, plotted over time. • Highlights — displays situations in which an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior. Managing Oracle Database Systems: User and Reference Guide 91 Chapter 2—Using Foglight for Oracle Positioning the cursor over a dimension type Upon positioning the cursor over a dimension type (for example, OS Users or DB Users), the lower section displays the following panes: • Top dimension section — includes a chart and a metrics table that display information about the top-consuming entities within the dimension selected on the left (for example, Top DB Users), the lower section contains the following panes: • Chart — displays the CPU wait for the dimension in the selected row. • Table for top resources — for details, see “Table for top resources (Redo panel)” on page 91. Table for top resources (Redo panel) The table displayed for top resources includes the following columns: • Actions — the actions that can be carried out for the selected resource within the selected row Note The actions that can take place when positioning the cursor on the top resource dimension and clicking the cogwheel button ( ) depend on whether Performance Analysis is installed and configured. For details, see “The Actions row” on page 65. • Resource type name — the name of the selected resource type; for example, database name or OS user name • Redo Wait — see “Redo Wait” on page 218. • SQL Executions — the number of SQL statements executed during the specified time range Managing Oracle Database Systems: User and Reference Guide 92 Chapter 2—Using Foglight for Oracle • Degree of Parallelism — the total number of server processes (master and slaves) assigned to handle a single operation. Cluster panel The contents of the Cluster panel vary between RAC and instance view, with and without Performance Analysis installed and configured, as described in the following sections: • “Cluster panel at the RAC level — without Performance Analysis” on page 92 • “Cluster panel at the RAC level — with Performance Analysis” on page 92 • “Cluster panel at the instance level — with Performance Analysis” on page 93 Cluster panel at the RAC level — without Performance Analysis The Whole Cluster mode contains the following sections: • Cluster workload — contains the Cluster Wait chart. For details, see “Cluster Wait chart” on page 92. • Global Cache Efficiency — for details, see “Global Cache Efficiency bar” on page 92. • Cluster Workload - Instance Breakdown — displays the cluster’s workload, divided by the various RAC nodes • Global Cache — contains a pie chart, which displays the distribution of the global cache among the RAC nodes, both in megabytes and in percents. • Cluster Balance — displays the balance between the various RAC nodes, divided by the following parameters: transfer rate, cache size, and latency. • Balance Criterion — a bar display of the criterion of the balance among the RAC nodes. This data is identical to the one displayed in the Cluster Balance section. Cluster Wait chart The Cluster wait chart displays the occurrence of the following wait event types over the specified time range, in either baseline (aggregated to the total Cluster Wait level) or breakdown (distribution between the various wait types) format: • Cluster coordination — the elapsed time spent while coordinating Real Application Cluster resources at cluster level • Network/Interconnect — wait events resulting from the communication link between the nodes • Global locks — locks that take place between the RAC’s nodes Global Cache Efficiency bar The Global Cache Efficiency bar displays the average cache efficiency of all RAC nodes. Cache efficiency measures, in percentages, the number of cache hits, that is, the number of times an instance submitted a request for the block and found it in the cache (either its own cache or the cache of another RAC node). The higher the rate of cache hits, the higher the cache efficiency value. Cluster panel at the RAC level — with Performance Analysis The Whole Cluster mode contains the following sections: Managing Oracle Database Systems: User and Reference Guide 93 Chapter 2—Using Foglight for Oracle • Cluster workload — contains the Cluster Wait chart. For details, see “Cluster Wait chart” on page 92. • Global Cache Efficiency — for details, see “Global Cache Efficiency bar” on page 92. • Statistics and highlights — includes the following panes: • Overview — contains the following sections: Metric table — currently includes only the Cluster Wait metrics, which indicates common cluster-wide membership node view encompassing cluster resources (such as interconnect hardware, shared disks, and Oracle instances). A chart, which displays the metric selected in the table. • Cluster Metrics — including the following sections: Cluster Workload - Instance Breakdown — displays the cluster’s workload, divided by the various RAC nodes Global Cache — contains a pie chart, which displays the distribution of the global cache among the RAC nodes, both in megabytes and in percents. Cluster Balance — displays the balance between the various RAC nodes, divided by the following parameters: transfer rate, cache size, and latency. Balance Criterion — a bar display of the criterion of the balance among the RAC nodes. This data is identical to the one displayed in the Cluster Balance section. • Highlights — displays situations in which an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior Cluster panel at the instance level — with Performance Analysis The Instance Cluster mode contains the following sections: • Cluster workload — contains the Cluster Wait chart. For details, see “Cluster Wait chart” on page 92. • Statistics and highlights — includes the following panes: • Overview — identical to the Overview pane at the cluster level • Interconnect Load — see “Sections displayed under the Interconnect Load pane” on page 93 • Interconnect Latency — see “Sections displayed under the Interconnect Latency pane” on page 94 • Highlights — displays situations in which an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior Sections displayed under the Interconnect Load pane The sections displayed under the Interconnect Load pane are: • Logical Reads — displays the number of logical reads carried out by the selected RAC node during the specified time range Managing Oracle Database Systems: User and Reference Guide 94 Chapter 2—Using Foglight for Oracle • Misses — displays the number of times the instance submitted a request for the block and did not find it in the cache (either its own cache or the cache of another RAC node) • Transfer Rate — displays the transfer rate of blocks during the specified time range, in either of the following views: • Baseline — shows the total interconnect load • Breakdown — shows the interconnect load, divided by current blocks and consistent read (CR) blocks • Spinners — this section contain spinners that displays the rate of the following operations: • Logical reads • Misses • Transfer This field displays real-time behavior, color-coded for severity. Sections displayed under the Interconnect Latency pane The sections displayed under the Interconnect Latency pane are: • Interconnect latency — displays the amount of time it takes to deliver a single data item from memory to a CPU over the interconnect network. The interconnect latency can be displayed in either of the following views: • Compare to Thresholds — comparing the latency experienced by the selected instance with the thresholds defined in the Cluster Latency collection. • Compare to Cluster Average — comparing the latency experienced by the selected instance with the average interconnect latency within the various cluster nodes. • CR Block Request Latency — the average latency for completing the round-trip time for the requests for consistent read blocks, shown using the following parameters: • Send latency — the time required by LMS to initiate a send of a consistent read block • Flush latency — the time it takes carry out a log flush when a CR request is served. The LMS must wait for the LGWR to complete flushing the changes to a buffer that is on the log flush queue, before this buffer can be sent. • Build latency — the time that the LMS process requires to create a consistent read block on the holding instance • Current Block Request Latency — the average latency for completing the round-trip time for the requests for current read blocks, shown using the following parameters: • Send latency — the time it takes to send the current block to the requesting instance over the interconnect network • Flush latency — the time it takes to flush to disk the changes that were made to a block (that is, to carry out a forced log flush), before the block is shipped to the requesting instance • Pin latency — the time required for pinning the current block before shipping it to the requesting instance. Pinning a block ensures that no further changes are made to the block while it is prepared to be shipped to another instance. Managing Oracle Database Systems: User and Reference Guide 95 Chapter 2—Using Foglight for Oracle Other panel Use the Other panel to monitor the time spent waiting for the completion of miscellaneous operations, that is, operations that cannot be classified into any other wait categories. This panel contains the following sections: • Workload — contains the Workload (Average Active Sessions) chart; see “The Workload chart” on page 95 • Statistics and Highlights section — includes the following panes: • Overview — contains the metric table and a chart that represents the metric selected in this table, that is, Other Wait. • Highlights — displays situations in which an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload, likely to be exceptional and significant behavior. The Workload chart The Workload (Average Active Sessions) chart displays the metrics listed in the table below, in either of the following manners: • A baseline value — displays only the Latch Wait metric • A breakdown — displays the other wait value as part of the total activity, by comparing it with the non-other wait The Workload (Average Active Sessions) chart displays metrics such as: • Miscellaneous • Non-other activities • Resource Manager Wait • Backup/ Recovery Wait Managing Oracle Database Systems: User and Reference Guide 96 Chapter 2—Using Foglight for Oracle Reviewing Overall Activity The Oracle Overview page provides an overall view of the RAC or instance status for the selected time range, through the views and tools listed below. A H View G View/Tool Name D C B F Explained in Section A Current Instance/ RAC activity “Monitoring Current Activity” on page 97. B Current Instance/ RAC storage status “Monitoring Current Storage” on page 98. C Indicators of the activity running on the instance “Tracking Throughput” on page 98. E For Further Details... Managing Oracle Database Systems: User and Reference Guide 97 Chapter 2—Using Foglight for Oracle View View/Tool Name Explained in Section D Historical RAC/ Instance activity “Monitoring Historical Activity” on page 99. E Performance Analysis configuration status “Reviewing the Status of the Connection to Performance Analysis” on page 100. E Alarms “Monitoring Alarms” on page 99 F Reports “Generating Reports” on page 101 G Activity Highlights “Reviewing Top Activityrelated Issues” on page 101 Note For Further Details... History Summary page — see “Reviewing RAC Historic Activity” on page 103 and “Reviewing Instance Historic Activity” on page 109. The ASM (Automatic Storage Management) and archiving components are enabled only when being used by the monitored instances. Monitoring the ASM-related data is possible only when the ASM instance is installed and configured. For details, see “Adding ASM Instances” on page 34. Monitoring Current Activity The Real-time Status pane provides the indicators listed below for monitoring the current activity of either the selected instance/RAC node or the entire RAC, in Whole Cluster view. • Sessions (no.) — displays the total number of currently running sessions, distributed to inactive sessions (that is, sessions that did not initiate an SQL call to Oracle), and active sessions. • CPU (%) — displays the percentage of CPU that was consumed on the Oracle host by all processes (Total) or by Oracle processes only (Oracle). • Memory (%) — displays the percentage of RAM that was consumed on the Oracle host by all processes (Total) or by Oracle processes only (Oracle) • Up Since — identifies the date and time when the instance/RAC node was last started. If the database is not running this area reads Database is down, and a popup is available, displaying the database status. Note In Whole Cluster view, this indicator displays data regarding the most recently started RAC node. Managing Oracle Database Systems: User and Reference Guide 98 Chapter 2—Using Foglight for Oracle • Response Time — the response time, in milliseconds, of the instance or RAC node. Note In Whole Cluster view, this indicator displays the maximum (slowest) response time among all of the cluster nodes. • Blocked Sessions (no.) — the total number of sessions waiting for a lock held by another session. • Avg. Virtual. Overhead — displays the percentage of CPU that is unavailable to this virtual machine because it is being utilized either by other virtual machines or by VMware itself. Note The virtualization overhead metric holds value only for virtual hosts running on VMware ESX servers, and had their VMware connection profile defined using the Databases Administration dashboard. For more information, see “Defining the VMware Connection Profile” on page 171. Clicking the average virtualization overhead indicator displays the Virtualization Resource Utilization popup. This popup shows the CPU usage by both the virtual machine and the ESX host, as well as the memory sharing and swapping between all powered-on virtual machines on the host. Monitoring Current Storage The Real-time Storage pane enables monitoring of the current storage utilization by means of the following indicators: • Total — displays the total amount of datafile storage available (used and free), in megabytes • Free — displays the amount of free storage as a percentage • Autoextensible — indicates if at least one datafile in the database is autoextensible: • No — none of the datafiles in the database are autoextensible • Yes — at least one datafile in the database is autoextensible • Archive Space — if archiving is configured, displays the following indicators of storagerelated information about the most critical archive destination, that is, the archive destination with shortest time to failure: • Cylinder — shows the amount of used archive destination memory • Time to Failure (hours) — indicates the estimated time to failure in hours, based upon the rate of archiving Important If the archive files reside on a disk that is managed by ASM, the Archive Space cylinder is disabled and the indication ASM agent is required appears. To retrieve archive data, the ASM instance should be installed and configured. For instructions, see “Adding ASM Instances” on page 34. Tracking Throughput The Real-time Throughput section enables tracking of the amount of activity running on the instance, using the speedometer charts displayed in the following sections: • Logons — the rate per second of sessions that were initiated during the specified time range • Executions — rate per second of statement executions during the specified time range Managing Oracle Database Systems: User and Reference Guide 99 Chapter 2—Using Foglight for Oracle • Physical Data Processed — the total size (MB) of data processed through physical read and write operations during the specified time range Monitoring Historical Activity The History pane indicators, described in the following sections, provide information for tracking the instance activity during the time range selected at the upper right corner of the screen: • “Workload Chart” on page 99 • “Top 3 busiest resources” on page 99 Workload Chart The Workload chart displays the workload of the average active sessions, plotted over time. The workload can be plotted as either: • Baseline — workload plotted over time • Breakdown — the amount of time the database spent waiting for various wait events, plotted over time To display the wait events breakdown, click the arrow to the right of the Breakdown link, as shown below. Top 3 busiest resources The Top 3 busiest resources section displays the three resources whose wait time constituted the highest percentage of the total active time during the specified time range. Monitoring Alarms The Alarms pane displays color-coded indicators of the monitored host status. The alarms are divided into the following types: • Baseline deviations — invoked for several of the metrics for which a baseline is configured • Thresholds — displays alarms about metrics or workloads, whose configured threshold has been exceeded When metrics or workload resources behave normally, without any deviations, the alarm representation is displayed in gray. When a configured baseline or threshold has been exceeded, alarms are displayed in colors that represent levels of severity, ranging from yellow to red. See the Foglight Online Help for details on the alarm information. Managing Oracle Database Systems: User and Reference Guide 100 Chapter 2—Using Foglight for Oracle The following indicators apply to all alarm types, including thresholds. Note The numbers displayed inside the indicators show the number of alarms encountered for each alarm type. Indicator Description The metric (or workload resource) has exceeded the configured threshold or deviated from the baseline. Yellow (Warning) The metric (or workload resource) has significantly exceeded a threshold or deviated greatly from the baseline. Orange (Critical) Red (Fatal) Note A severe connectivity issue was encountered, thereby triggering a fatal alarm. Such an issue can occur in one or more of the components that take place in the monitoring process; for example, if Oracle database is not responding, the listener is down, or the number of connections reached the number of sessions defined in the init.ora file. It is possible to suspend normal monitoring activities (data collection or alarms) due to administrative preference, by defining a blackout period. Blackouts are commonly created to prevent noisy alerts during scheduled maintenance periods. For details, see “Creating Blackouts for Selected Instances” on page 102. Enabling or disabling baseline calculation The Enable/Disable Baseline command under the Databases Technical Support dashboard allows enabling or disabling the baseline calculation for selected metrics. To enable or disable baseline calculation: 1 Go to Dashboards > Databases > Support Dashboard > Databases Technical Support. 2 Click near the Enable/Disable Baseline command. 3 Select the Oracle tab. 4 Select the requested metrics under the Metric Name column. 5 Select the requested command, either Enable Baseline or Disable Baseline. Reviewing the Status of the Connection to Performance Analysis The Performance Analysis configuration indicator notifies if Performance Analysis is installed and configured for the selected instance. Managing Oracle Database Systems: User and Reference Guide 101 Chapter 2—Using Foglight for Oracle If Performance Analysis is installed, this application can be launched from any screen using the In-context actions button. For more information, see “Components Shared by All Foglight for Oracle Screens” on page 50. If Performance Analysis is not installed, click the status link (for example, Performance Analysis is not installed) to open the Performance Analysis screen, used for maintaining the connection to Performance Analysis. For more information, see “Configuring the Connection to Performance Analysis” on page 193. The possible states of the Performance Analysis configuration indicator are: • Powered by Performance Analysis • Performance Analysis is not installed • Performance Analysis connection failed • Performance Analysis agent not running • Performance Analysis not supported • Performance Analysis partially supported • Performance Analysis upgrade recommended Generating Reports The Reports section can be used for generating the following types of reports: • Reports about various aspects of the selected instance or RAC performance. These are identical to the reports generated using the Dashboards > Report command. For information on how to generate the various reports, as well as a brief description of each report, see “Generating Reports” on page 199. • Executive summary report — a one-click report-generating feature, designed to deliver the critical parameters displayed on the Overview Page in a useful configurable format to key decision-makers. If the monitored instance is part of a multitenant architecture, this report displays a grid showing data for each pluggable database. Reviewing Top Activity-related Issues The Activity Highlights section, which is active only when Performance Analysis is installed and configured, provides access to the Performance > Highlights pane. This pane lists situations in which an individual or small activity subset (characterized by dimension values) is the primary source of an activity (resource) workload which are likely to be exceptional and significant behavior. Managing Oracle Database Systems: User and Reference Guide 102 Chapter 2—Using Foglight for Oracle Generating Reports Creating Blackouts for Selected Instances In certain scenarios, users may prefer not to receive alarms from one or more agents for a predefined period of time (for example: during the night hours or when a scheduled maintenance is carried out). Creating a blackout for one or more databases within an instance requires defining the requested instance as a service, and then creating a blackout for this service. To create a service: 1 Go to Dashboards > Services > Service Builder. 2 Click the sign near the title Add a New Category. The New Category Wizard dialog box opens. 3 Enter a name for the service in the Name field. 4 In the Tier list, select the User option. 5 Select the option No in the field that requests indicating whether Foglight should dynamically add and maintain the hosts for this service. 6 Click Finish. Managing Oracle Database Systems: User and Reference Guide 103 Chapter 2—Using Foglight for Oracle The newly created service appears on the table. 7 In the Actions column, click Add ( ) and select the command Add components to this service. The dialog box Add components to this service opens. 8 Click Add specific component. 9 Use the From list to select the option Agents. 10 Click Search. The list of agents appears on the table. 11 Select the boxes near the requested agents. 12 Click Add Components. 13 Repeat step 7 to step 8. 14 Use the From list to select the option All Models. 15 Click the + sign near the name DBSS_Data_Model, to expand this field. 16 Select the instances for which no alarms should be invoked. 17 Click Add Components. To create a blackout for the newly created service: 1 Go to Dashboards > Administration > Setup & Support > Blackouts. 2 Click the option Create a One-Time Blackout. 3 In the next screen select the option Suspend Alarms. 4 In the screen Choose Target Type, select Services and click Next. The screen Choose Blackout Targets appears. 5 Select the previously created service for which no alarms are to be invoked and click Next. 6 Use the next screen, Specify Blackout Period, to select the start and end times of the blackout period and click Next. The screen Configure Name and Reason appears. 7 Select whether to leave the default name of the blackout or to enter a new name, as well as the reason for blackout. 8 Click Finish to complete the procedure. The Blackout Summary screen appears, displaying the details of the newly created blackout. Reviewing RAC Historic Activity The History Summary page in Whole Cluster view contains the following panes: • Cluster Overview • Interconnect/GCS • Storage Managing Oracle Database Systems: User and Reference Guide 104 Chapter 2—Using Foglight for Oracle • I/O Activity • Resource Utilization • Sessions In addition, at the bottom of the page, Foglight for Oracle indicates whether Performance Analysis is connected. For details, see “Reviewing the Status of the Connection to Performance Analysis” on page 100. Monitoring the Cluster’s Usability and Load Balance The Cluster Overview pane displays the cluster’s overall functioning and load balance. Cluster Section • Availability — the percentage of RAC nodes that are available (up) and running. Clicking the number on the icon displays a popup that shows the percentage of cluster nodes that were available (up) during the specified time range, plotted over time. Click this indicator to drill down to the Activity > Cluster Usability panel. For details, see “Reviewing Cluster Usability” on page 124. • Response Time — the maximum (slowest) response time among all of the cluster nodes, in milliseconds. Holding the cursor over the number on the icon displays a dwell that shows the average response time, that is, the average time (in milliseconds) that elapses from the moment a dummy query (either select 1 from dual or a user-defined query) is submitted, until the application indicates that the query was executed. This value is plotted over time. Click this indicator to drill down to the Activity > Cluster Usability panel. For details, see “Reviewing Cluster Usability” on page 124. Managing Oracle Database Systems: User and Reference Guide 105 Chapter 2—Using Foglight for Oracle Balance Criterion Section The Balance Criterion section contains the following charts: • Values aggregated to the cluster level • Bar chart values for each cluster node These charts display data according to the value selected from the Balance Criterion list. The available values are: • Workload — in average active sessions • Availability — average availability, in percent • Response Time — maximum response time, in milliseconds • Logical Reads — average number of logical reads, in blocks • CPU Load — average CPU load, in percent • Interconnect Load — average interconnect load, in blocks • Cluster Wait Time — average cluster wait time Balance Section The Balance section contains the balance bars listed below, which display the balance rate of RAC nodes for each metric. • Workload (%) — the extent, in percent, to which workload (average active sessions) is balanced across all RAC nodes. Ideally, all instances should equally share the wait time caused by active sessions workload; however, this metric can indicate a situation where a specific instance takes a much larger share, thereby possibly overloading the system. • I/C Load (%) — the extent, in percent, to which the Interconnect workload-related wait events are balanced across all RAC nodes. Ideally, all instances should equally share the wait time caused by Interconnect workload; however, this metric can indicate a situation where a specific instance takes a much larger share, thereby possibly overloading the system. • Cluster Wait (%) — the extent, in percent, to which Cluster wait is balanced across all RAC nodes. Ideally, all instances should equally share the wait time caused by cluster wait; however, this metric can indicate a situation where a specific instance takes a much larger share, thereby possibly overloading the system Click the number on the icon to display a popup that shows the percentage of cluster balance between all cluster instances. This value is plotted over time. • Logical Reads (%) — the extent, in percent, to which Logical Reads are balanced across all RAC nodes. Ideally, all instances should equally share the wait time caused by logical reads; however, this metric can indicate a situation where a specific instance takes a much larger share, thereby possibly overloading the system. Managing Oracle Database Systems: User and Reference Guide 106 Chapter 2—Using Foglight for Oracle • CPU Load (%) — the extent, in percent, to which CPU usage is balanced across all RAC nodes. Ideally, all instances should equally share the wait time caused by CPU usage; however, an overly low value of this metric can indicate a situation where a specific instance takes a much larger share, thereby possibly overloading the system Monitoring Interconnect, Cluster, and Global Cache The Interconnect/GCS pane contains the indicators listed below, which are used for monitoring the global cache functioning. These indicators also demonstrate the effect of the global cache functioning over the interconnect load and the cluster latency and overhead. • Cluster Latency — the average cluster latency value, in milliseconds. Cluster latency is the time that has elapsed since the request for the block was submitted until the block was received by the requesting instance. • Cluster Overhead — the percentage of cluster wait events within the overall active time. This value summarizes the cluster-related activity of all cluster instances. • Global Cache — the total amount of global cache in megabytes. This value represents the amount of space allocated to Global Cache across all instances in the cluster. • Cache Miss Rate — the maximum cache miss rate in percent. This value represents the frequency at which a data block is requested from a remote instance, after the user session failed to find this block in the buffer cache. • I/C Load — the total Interconnect Load across all cluster nodes, that is, the aggregated number of requests for blocks that were not satisfied from the local instance, and therefore had to make a request for the block from the remote instance. Monitoring Disk Storage The Storage pane contains the components listed below, which display storage status. The information shown in this pane is aggregated to the RAC level. Datafiles-related Indicators • Cylinder — displays the total amount of used and free datafile storage. Clicking the cylinder displays a popup, which shows the total space usage by the database, plotted over time. This popup can also be used to drill down for further investigation, using the Tablespace Storage drilldown. For further details, see “Tablespaces page” on page 138. • Total — displays the total amount of datafile storage available (used and free), in megabytes. • Free — displays the amount of free RAC storage, in percent. • Autoextensible — indicates if at least one datafile in the database is autoextensible: • No — all of the datafiles in the database are not autoextensible. • Yes — at least one datafile in the database is autoextensible. Managing Oracle Database Systems: User and Reference Guide 107 Chapter 2—Using Foglight for Oracle Archive Destinations-related Indicators If archiving is configured, the following indicators under the Archive Space (MB) section display storage-related information about the most critical archive destination, that is: the archive destination with shortest time to failure. This section contains the following indicators: • Cylinder — shows the amount of used archive destination memory. • Time to Failure (hours) — indicates the estimated time to failure in hours, based upon the rate of archiving. Important If the archive files reside on a disk that is managed by ASM, the Archive Space cylinder is disabled and the indication ASM agent is required appears. To retrieve archive data, the ASM instance should be installed and configured. For instructions, see “Adding ASM Instances” on page 34. Indicators of Storage-related ASM Information If Automatic Storage Management (ASM) is configured, and the ASM instance is installed and configured, the following indicators display storage-related information about ASM: • Cylinder — displays the total amount of allocated and free ASM storage memory. Clicking the cylinder displays a popup, which shows the space usage of storage managed by ASM, plotted over time. This popup can also be used to drill down for further investigation, using the Tablespace Storage drilldown. For further details, see “Tablespaces page” on page 138. • Allocated (megabytes) — the total amount of ASM memory available (allocated and free), in megabytes. • Free (percent) — the amount of free RAC ASM storage memory, in percent. Monitoring the Cluster’s I/O Activity The Average I/O activity pane uses the indicators listed below to display I/O status information, aggregated to the RAC level. • Block Read Service Time — the maximum amount of service time taken for a block read, in milliseconds. Click the number on the icon to display a popup that shows the amount of time required for reading a single block into memory from a datafile on each instance, plotted over time. This indicator can also be used for further investigation by means of the IO Activity for RAC drilldown. For details, see “Disk I/O panel in Whole Cluster mode” on page 69. • Physical Reads — the average number of physical reads, in blocks per second. Click the number on the icon to display a popup that shows the rate at which data blocks are read from disk on all of the cluster nodes. This value is plotted over time. This indicator can also be used for further investigation by means of the IO Activity for RAC drilldown. For details, see “Disk I/O panel in Whole Cluster mode” on page 69. Managing Oracle Database Systems: User and Reference Guide 108 Chapter 2—Using Foglight for Oracle • Physical Writes — the average number of physical writes, in blocks per second. Click the number on the icon to display a popup that shows the rate at which modified blocks are written from the SGA to disk by the DBWR processes on all of the cluster nodes. This value is plotted over time. This indicator can also be used for further investigation by means of the IO Activity for RAC drilldown. For details, see “Disk I/O panel in Whole Cluster mode” on page 69. • Redo Writes — the average number of redo writes, in writes per second. Click the number on the icon to display a popup that shows the rate at which redo log entries are written to the redo log files by the LGWR processes on all of the cluster nodes. This value is plotted over time. This indicator can also be used for further investigation by means of the Redo and Archive Performance for RAC drilldown. For further details, see “Redo panel at the RAC level” on page 88. The following indicators display data only if Automatic Storage Management (ASM) is configured, and the ASM instance is installed and configured (see “Adding ASM Instances” on page 34): • ASM Reads — the total number of reads carried out by ASM disk groups. Click the number on the icon to display a popup that shows the total reads carried out by the ASM disk groups, plotted over time. • ASM Writes — the total number of writes carried out by ASM disk groups. Click the number on the icon to display a popup that shows the total reads carried out by the ASM disk groups, plotted over time. Monitoring Resource Utilization Use the Resource Utilization pane to perform the tasks described in the following sections: • Monitoring the CPU Load • Monitoring the RAM Monitoring the CPU Load The indicators listed below are used for monitoring the CPU load. • CPU (%) — the average CPU load (percentage), during the specified time range, of all CPU units that host the RAC nodes. This indicator displays the distribution between Oracleincurred load and total CPU load. • Number of CPUs — the average number of CPU units on all of the RAC nodes’ hosts. Clicking the number on the icon displays a popup that shows the total CPU usage on the cluster nodes, plotted over time. Monitoring the RAM The indicators listed below are used for monitoring the RAM consumption. Managing Oracle Database Systems: User and Reference Guide 109 Chapter 2—Using Foglight for Oracle • Memory (%) — the average memory consumption (percentage), during the specified time range, of all CPU units that host the RAC nodes. This indicator displays the distribution between Oracle-incurred memory consumption and total memory consumption. • Total RAM — the total amount of RAM configured for the RAC database nodes, in megabytes. • Free RAM — displays the percentage of RAM not in use by the cluster nodes. Monitoring Session Activity Use the Sessions section to monitor the session activity that took place during the specified time range, by comparing between the following indicators: • Avg. Active Sessions — the average number of active sessions that experienced resource contention during the selected time range. • Avg. Connected Sessions — the average number of sessions that connected to the database during the selected time range. Reviewing Instance Historic Activity The Oracle Database instance History Summary page provides a summary view of the instance health during the specified time range, and can be used to drill down even further for carrying out root cause analysis of various performance issues. Note Archive and Data Guard data is available only if these modules have been configured. ASM data is available only storage management is carried out by ASM and the ASM instance is installed and configured. Managing Oracle Database Systems: User and Reference Guide 110 Chapter 2—Using Foglight for Oracle This dashboard contains the following sections: • Workload • Storage • Cache Hit Ratio • Data Guard • Availability • OS Performance Monitoring Workload Use the Workload pane to track the workload of the average active sessions, using the Workload chart, which displays the workload of the average active sessions, plotted over time. The workload can be plotted as either: • Baseline — workload plotted over time. • Breakdown — the amount of time the database spent waiting for various wait events, plotted over time. Note The chart and the indicators below do not display all wait event types, but only the top wait events, that is, the wait events that consumed the highest amount of active time during the selected time range. Monitoring Disk Storage The Disk Storage panel contains the following components: Managing Oracle Database Systems: User and Reference Guide 111 Chapter 2—Using Foglight for Oracle • Storage chart — displays the following indicators regarding the datafiles storage: • Incremental change growth — the growth of the size occupied by the database, plotted over the specified time range. • Database allocated size • Avg. Consumption — the average rate (MB/hour) of disk space allocated to the database • Total Allocated — the total amount of space allocated to the database • ASM — indicates whether the database is monitored using automatic storage management (ASM). Important Starting from version 5.6.4.32x, the ASM-related data that was previously retrieved through the Foglight for Oracle agent is only enabled if the Foglight ASM instance is configured and running. For instructions about installing the ASM instance, see “Adding ASM Instances” on page 34. Viewing the Instance Cache Hit Ratio The Cache Hit Ratio section displays the percentage of file read operations that were satisfied by the file system cache without requiring any physical I/O. The value of this metric should be as high as possible. The cache hit ratio is displayed for the following cache memory types: • PGA — the percentage of work areas executed by the system, which processed the entire input data exclusively using the PGA memory • Buffer cache — the percentage of block requests, which found the block in the buffer cache, out of the entire block requests (including requests that required disk reads) • Library cache — the percentage of queries that found the parsed execution plan in the library cache (soft parsing) The information is provided by both a text and an icon, which displays real-time behavior, colorcoded for severity. Clicking the icon displays a popup that shows the cache hit ratio in a chart form, plotted over the selected time range. This popup can also be used for further investigation, by means of the Performance drilldown. For further details, see “Reviewing Performance-related Data” on page 59. Monitoring the Data Guard Service The Data Guard section enables the monitoring of Oracle Data Guard on primary databases, by displaying the gap between the last archive log on the primary database and last applied archive log on the standby database. Because Oracle Data Guard supports physical standby databases that use Redo Apply technology, use this section to monitor the rate of applying redo log files on standby database instance. Note The Data Guard section displays data only if the Data Guard service is configured. Managing Oracle Database Systems: User and Reference Guide 112 Chapter 2—Using Foglight for Oracle Monitoring the Instance’s Availability The indicators included in the Instance pane provide various details about the instance. All of the indicators display real-time behavior, color-coded for severity. Clicking each indicator displays a popup that shows the metric’s value, plotted over time. These indicators are: • Sessions — displays a chart showing the total number of sessions, distributed between the following session types: • Avg. Active Sessions — the average number of active sessions that experienced resource contention during the selected time range. • Avg. Connected Sessions — the average number of sessions that connected to the database during the selected time range. • Availability — the average database instance availability, in percent, that is, the percentage of time that the instance was available during the selected time range. This value is displayed as a number and a bar gauge. Click this indicator to drill down to the Activity > Instance Usability panel. For details, see “Reviewing Instance Usability” on page 123. • Average Response Time — the average response time, in milliseconds, that elapsed from the moment a dummy query (either select 1 from dual or a user-defined query) is submitted, until the application indicated that the query was executed. This value is plotted over time. Click this indicator to display a popup that shows the actual response time, plotted over the specified time range. This popup can also be used to drill down to the Activity > Instance Usability panel. For details, see “Reviewing Instance Usability” on page 123. Monitoring the Host Activity The indicators displayed on the Host Activity pane provide various details about the host activity, that is, the host’s CPU usage and memory consumption, affected by all of the processes rather than Oracle processes only, as well as the system I/O load. All of the indicators display real-time behavior, color-coded for severity. Clicking each indicator displays a popup that shows the metric’s value, plotted over time. These indicators are: • Average CPU (%) — The overall operating system CPU usage, including Oracle • Average Memory (%) — The percentage of memory consumed by all of the instance’s OS processes (including Oracle), compared with the maximum amount of memory the instance could use. This metric includes both resident and swapped RAM. • Disk Utilization (%) — the percentage of time the busiest device spent serving system-wide I/O requests. This metric serves as a measure for the system I/O load. Managing Oracle Database Systems: User and Reference Guide 113 Chapter 2—Using Foglight for Oracle Viewing the Foglight for Oracle Real-time Summary Page The following image displays the main elements of the Foglight for Oracle real-time summary page. The table that follows the image identifies each element, and provides a link to display more information, corresponding to the logical dataflow within Foglight for Oracle Instance Homepage. A - Instance properties C - DataGuard service J - Performance Analysis configuration indication G - Disk storage D - Workload E - Process activity F - Memory activity H - In-context actions button I - Physical I/O Operations B - Sessions Tool/View Tool/View Name Explained in Section A Instance properties “Identifying the Instance” on page 114 “Monitoring the Instance” on page 116 B Sessions pane “Monitoring General Sessions’ Statistics” on page 116 C DataGuard pane “Monitoring the Data Guard Service” on page 117 D Workload “Monitoring the Workload” on page 118 E Process activity “Monitoring Process Activity” on page 119 Managing Oracle Database Systems: User and Reference Guide 114 Chapter 2—Using Foglight for Oracle Tool/View Tool/View Name Explained in Section F Memory Activity pane “Monitoring Memory Activity” on page 119 G Disk Storage pane “Monitoring Disk Storage” on page 120 H In-context actions button “Components Shared by All Foglight for Oracle Screens” on page 50 I Physical I/O operations “Tracking Physical I/O Activity” on page 121 J Performance Analysis configuration indication “Reviewing the Status of the Connection to Performance Analysis” on page 122 The main groups (panes) of gathered icons and gauges, which represent the main activity areas in the Oracle connection process, are as follows: • Instance identification — used for identifying the instance, its type, and its properties. See “Identifying the Instance” on page 114. • Components representing instance data flow — the main activity area in the Foglight for Oracle Instance Homepage includes several panes and flows that represent the data flow in the Oracle operation. The dataflow is represented in a top-down design, that is, from the session to the physical disk storage. Identifying the Instance Use the Instance identification indicators to identify the instance, its type and its operation period. Managing Oracle Database Systems: User and Reference Guide 115 Chapter 2—Using Foglight for Oracle Instance Name Specified time range Instance pane These indicators are: • Instance name — identifies the instance. • Specified time range — indicates the period of time for which data is being displayed (by default: last 60 minutes). • Instance pane — contains the following indicators: • DB Type — identifies whether the monitored database’s type is SQL Server, Sybase, DB2 or Oracle. • DB Version — identifies the Oracle version number, along with the most recently installed service packs. • Up Since — identifies the date and time when the instance was last started. Note The pane’s initial view only shows the date; to display the time as well, hover the mouse or click the date text. • OS Version — identifies the operating system’s version number, along with the most recently installed builds and service packs. Note The pane’s initial view only shows the OS name; to display the build and service packs as well, hover the mouse or click the OS name’s text. Managing Oracle Database Systems: User and Reference Guide 116 Chapter 2—Using Foglight for Oracle Monitoring the Instance The indicators included in the Instance pane provide various details about the instance. All of the indicators display real-time behavior, color-coded for severity. Note Clicking the Response Time, CPU (%) and Memory (%) indicators displays a popup that shows the metric’s value, plotted over time. • Response Time — the time (in milliseconds) that elapses from the moment a query, which is supposed to represent the general workload, is submitted, until the application indicates that the query was executed. As the response time is usually the starting point for investigation, the Response parameter leads to the SQL Instance Summary panel in the SQL Activity drilldown, • CPU (%) and Memory (%) — the average CPU load and memory consumption (percentage), during the specified time range, of all CPU units that host the Oracle instance. This indicator displays the share of Oracle-incurred CPU load and memory consumption within the total figure. Clicking the number on both icons displays a popup that shows the total CPU usage or memory consumption on the host, plotted over time. • Number of CPUs — the number of CPU units on the instance’s host. • Total RAM — the total amount (in megabytes) of the host’s physical memory. • Free RAM — the total amount (in megabytes) of physical memory available to the applications Monitoring General Sessions’ Statistics The Sessions pane monitors all session types, that is: system, user, and Oracle sessions. Use this pane to view the response time compared with the number of sessions and the instance’s level of activity. A high response time value may result from a long queue, that is: an overly high percentage of active users. Long queues can indicate one of the following issues: • A massive workload — the system handles more users than it was initially designed to do. • A bottleneck — lack of system resources prevents users from carrying out their transactions, resulting in wait events and an increasing number of sessions that remain active for prolonged periods. The Sessions pane displays the total number of sessions, distributed according to the following distinctions: • Active vs. inactive sessions • System (background) vs. User (foreground) sessions The parameters in this pane lead to the Activity > Sessions panel. For details, see “Viewing Detailed Sessions Data” on page 129. The Client applications represented graphically in the Sessions pane communicate with Oracle by sending and receiving network packets and by submitting SQL statements for execution by Oracle. Managing Oracle Database Systems: User and Reference Guide 117 Chapter 2—Using Foglight for Oracle The flows, described in the following table, help indicate performance issues if their values are too low. Flow Description Executions The rate per second of calls (both user and recursive) that executed SQL statements. Received packets The rate at which network packets are being received by Oracle from Client applications. When this icon’s color is yellow, clicking it displays text that describes the deviations that triggered this display, along with a representation of the rate at which Oracle is encountering network packet errors. Sent packets The rate at which network packets are being sent from Oracle to Client applications. When this icon’s color is yellow, clicking it displays text that describes the deviations that triggered this display, along with a representation of the rate at which Oracle is encountering network packet errors. Monitoring the Data Guard Service Use the Data Guard section for monitoring Oracle Data Guard on primary databases. This section displays the gap between the last archive log on the primary database and last applied archive log on the standby database. Oracle Data Guard can be configured as either physical or logical standby. For details about the preparations required for monitoring each of these setups, see “Monitoring the Dataguard physical standby” on page 117 and “Working with logical standby (SQL Apply)” on page 118. As Oracle Data Guard supports physical standby databases that use Redo Apply technology, use this section to monitor the rate of applying redo log files on standby database instance. Important The Data Guard section displays data only if the Data Guard service is configured. Note Starting from Oracle 11g, Oracle provides the Active Data Guard option, which allows the standby to be read-only mode and at the same time to receive changes received from the primary node. Monitoring the Dataguard physical standby To prepare the physical standby: 1 Ensure that the Standby Database is in receiving mode. 2 Use sqlplus to connect to the Primary Database as SYS user. 3 Create the standby Foglight schema owner. Managing Oracle Database Systems: User and Reference Guide 118 Chapter 2—Using Foglight for Oracle 4 Run the privileges granting script for granting privileges to the created standby user, still from primary database. This script can be downloaded from the Connection Details screen. For details, see “Defining Connection Details” on page 166. If the Data Guard mode being used is Active Data Guard, proceed to step 6. Otherwise proceed to the next step. 5 Switch to read-only mode on the standby database. Note Switching to read-only is only for the standby user availability check; there is no need to stay in that mode after the checkup. 6 Ensure that the Foglight schema owner is available on the standby database. 7 Open a browser and run the Oracle Monitoring Installer wizard. 8 Select Add a New Instance. 9 Specify the standby database credentials and connection details (the user that was created in step 4). 10 Click Verify Connectivity After connectivity verification is completed successfully, start monitoring the physical standby database. Working with logical standby (SQL Apply) To prepare the logical standby: 1 Use sqlplus to connect to the Logical standby as SYS user. 2 Create the standby Foglight schema owner 3 Open a browser and run the Oracle Monitoring Installer wizard. 4 Continue the Oracle Monitoring Installer process as usual, using the Logical standby user credentials and connection details After connectivity verification is completed successfully, start monitoring the logical standby database. Monitoring the Workload The Workload chart displays the workload of the average active sessions, plotted over time. The workload can be plotted as either: • Baseline — workload plotted over time. • Breakdown — the amount of time the database spent waiting for various wait events, plotted over time. To display the wait events breakdown, click the arrow to the right of the Breakdown link, as shown below. Managing Oracle Database Systems: User and Reference Guide 119 Chapter 2—Using Foglight for Oracle Monitoring Process Activity The Process Activity pane enables the monitoring of the processes ran by the Oracle instance, using performance indicators such as the total number of blocked processes and the rate per second of parses that are carried out during the most recent sample. The Process Activity pane displays the parameters described below: Parameter Description Recent Executions The number of statements executed during the most recent sample. Blocked Processes The average number of blocked processes that were collected during the specified time range. Blocked processes are processes that are waiting for another process to release a resource that the process is currently locking. Such processes can sometimes result in bottlenecks. The Blocked Process indicator changes its color when one or more processes become blocked. Parses Total number of parse calls. This figure is the total of both hard parse calls (calls that require loading the SQL source code into RAM for parsing) and soft parse calls. Concurrent Transactions The number of currently running transactions. For further details, see “Oracle Activity Drilldown” on page 122. To communicate with the Oracle memory, the Oracle processes use logical reads and soft and hard parses, represented as flows in the homepage. Monitoring Memory Activity The Memory Activity session provides a view on the Oracle instance’s memory utilization, as described in the following table. Managing Oracle Database Systems: User and Reference Guide 120 Chapter 2—Using Foglight for Oracle Section Metric Description SGA Total (MB) The actual size currently allocated to the System Global Area (SGA) memory. Buffer Cache (MB) The amount of memory currently allocated to the buffer cache. The buffer cache, which is used for caching frequently accessed data blocks, is normally the largest memory cache used by Oracle. Shared Pool (MB) The size of the shared pool. The shared pool caches SQL statements, PL/SQL programs, object definitions, and session memory for MTS sessions. Performance can improve when the shared pool is sized properly, thereby avoiding re-parsing and reloading. Buffer Cache Hit (%) The percentage of block requests, which found the block in the buffer cache, out of the entire block requests (including requests that required disk reads). Library Cache Hit (%) PGA The effectiveness of the library cache, computed by comparing the number of queries submitted to Oracle server for execution, which found the parsed execution plan in the library cache (soft parsing), with the entire number of queries submitted during the specified time range (including queries that required the parsing of a new statement, also known as “hard parsing”). Total (MB) The actual size currently allocated to the Program Global Area (PGA) memory. PGA Hit (%) The percentage PGA cache hit ratio. When the value of this metric is 100%, it indicates that all work areas executed by the system processed the entire input data using exclusively the PGA memory, thereby making optimal use of this memory. Any lower value indicates that at least one work area required one or more extra passes over the input data. Monitoring Disk Storage The Storage pane contains the components listed below, which display storage status. The information shown in this pane is aggregated to the RAC level. Datafiles-related Indicators • Cylinder — displays the total amount of used and free datafile storage. Click the cylinder to drill down. A popup appears, displaying the total space usage by the database, plotted over time. Managing Oracle Database Systems: User and Reference Guide 121 Chapter 2—Using Foglight for Oracle This popup can also be used for further investigation by means of the Tablespace Storage drilldown. For further details, see “Tablespaces page” on page 138. • Total — displays the total amount of datafile storage available (used and free), in megabytes. • Free — displays the amount of free RAC storage, in percent. • Autoextensible — indicates if at least one datafile in the database is autoextensible: • No — all of the datafiles in the database are not autoextensible. • Yes — at least one datafile in the database is autoextensible. Archive Destinations-related Indicators If archiving is configured, the following indicators display storage-related information about the most critical archive destination, that is: the archive destination with shortest time to failure. This section contains the following indicators: • Cylinder — shows the amount of used archive destination memory. • Time to Failure (hours) — indicates the estimated time to failure in hours, based upon the rate of archiving. Indicators of Storage-related ASM Information If storage is managed using Automatic Storage Management (ASM), and the ASM instance is installed and configured, the following indicators display storage-related information about ASM: • Cylinder — displays the total amount of allocated and free ASM storage memory. • Allocated (megabytes) — the total amount of ASM memory available (allocated and free), in megabytes. • Free (percent) — the amount of free RAC ASM storage memory, in percent. Monitoring Disk Activity The Disk Activity section displays the following indicators: • Unarchived — if archiving is configured, indicates the amount of Redo Logs waiting to be archived, in percent. The popup that appears when clicking the Unarchived flow can also be used for drilling down further to investigate, using the “Redo Logs” drilldown. For further details, see “Redo Logs Panel” on page 146. • ASM I/O — a textual indication of the rate of ASM memory I/O operations. If storage is managed using Automatic Storage Management (ASM), and the ASM instance is installed and configured, this field displays real-time behavior, color-coded for severity. Tracking Physical I/O Activity The real-time summary page contains flows and graphic images that are used for tracking physical I/O activity, as listed below. All of the indicators display real-time behavior, color-coded for severity. Clicking each indicator displays a popup, which shows the metric’s value, plotted over time. Managing Oracle Database Systems: User and Reference Guide 122 Chapter 2—Using Foglight for Oracle • Physical Writes — displays the rate per second of physical writes, that is: the rate at which modified blocks are written from the SGA to disk by the DBWR processes. • Physical Reads — displays the rate per second at which data blocks are read from disk, when a connection requests a page that does not already exist in the buffer cache. • Redo Writes — displays the rate per second of redo writes, that is: the rate at which redo log entries are written to the redo log files by the LGWR processes. Reviewing the Status of the Connection to Performance Analysis The Performance Analysis configuration indicator notifies if Performance Analysis is installed and configured for the selected instance. Performance Analysis is a Quest technology, designed to collect SQL statements and performance indicators from Oracle databases and host operating systems, without imposing overhead on the target database itself. Performance Analysis is available only if Performance Analysis is installed. Performance Analysis has a server-based agent (Performance Analysis Agent), which monitors databases and workloads of all scales. If Performance Analysis is installed, this application can be launched from any screen by clicking the In-context actions button. For more information, see “Components Shared by All Foglight for Oracle Screens” on page 50. The possible states of this indicator are: • Powered by Performance Analysis • Performance Analysis not configured • Performance Analysis error • Performance Analysis is down Oracle Activity Drilldown Use the Oracle Activity drilldown to carry out the operations described in the following sections: • Reviewing usability and availability — using the Usability panel, either for a single instance or a RAC. For details, see “Reviewing Instance Usability” on page 123 and “Reviewing Cluster Usability” on page 124. • Viewing current and historical locks-related data — using the Locks panel, whose display changes when selecting a single instance or a RAC. For details, see “Reviewing Locks Activity at the Cluster Level” on page 125 and “Reviewing Locks Activity at the Instance Level” on page 128. • Viewing information about sessions that experienced the highest contention of a specified contention criterion — using the Sessions panel. For details, see “Viewing Detailed Sessions Data” on page 129. • Viewing I/O activity of all datafiles on the cluster or instance — using the panel Overall Datafiles I/O, whose display changes when selecting a single instance or a RAC. For details, Managing Oracle Database Systems: User and Reference Guide 123 Chapter 2—Using Foglight for Oracle see “Viewing Datafiles I/O Activity at the Cluster Level” on page 133 and “Viewing Datafiles I/O Activity at the Instance Level” on page 134. • Viewing the user-defined collections, which were created from the Databases Administration dashboard using the User-defined Collections screen. For details, see “Viewing UserDefined Collections” on page 134. Reviewing Instance Usability The Instance Usability panel displays aggregated information regarding the instance’s usability and availability. Instance pane Alarms pane Instance availability section Listener availability section Connection Time section Response Time section This panel supports the following tasks: • Identifying instances. For details, see “The Instance Pane” on page 124. • Viewing usability-related alarms. For details, see “The Alarms Pane” on page 124. • Tracking the instance’s availability, plotted over the defined time range. For details, see “The Instance Availability Section” on page 124. • Tracking the listener’s availability, plotted over the defined time range. For details, see “The Listener Availability Section” on page 124. • Tracking the instance’s connection time, plotted over the defined time range. For details, see “The Connection Time Section” on page 124 • Tracking the instance’s response time, plotted over the defined time range. For details, see “The Response Time Section” on page 124. Managing Oracle Database Systems: User and Reference Guide 124 Chapter 2—Using Foglight for Oracle The Instance Pane The Instance pane contains the following indicators: • DB Type — identifies monitored database’s type (for example: SQL Server or Oracle). • Oracle Version — identifies the Oracle version number, along with the most recently installed patches. • Up Since — identifies the date and time when the instance was last started. If the database is not running this area reads “Database is down”, and a popup is available, displaying the database status. The Alarms Pane The Alarms pane displays the usability-related outstanding alarms. These alarms alert the following issues per instance: response time, instance availability, instance connection availability, OS connection availability, instance connection time, and Listener status. The Instance Availability Section The Instance Availability section displays the instance availability in percentages, plotted over the time range. If the instance was unavailable once or more during this time, the relevant message describing the last occurrence of unavailability appears in the Error Message box. The Listener Availability Section The Listener Availability section displays the availability of all listeners configured for the instance, in percentages, plotted over the time range. If one or more of the listeners were unavailable during this time, a message describing this issue appears under the Listener Errors section. The Connection Time Section The Connection Time section displays the instance’s connection time, in seconds, plotted over the time range. The Response Time Section The Response Time section displays the instance’s response time, in milliseconds, plotted over the time range. Reviewing Cluster Usability Use the Cluster Usability panel to compare between the various RAC nodes, regarding the following indicators: • Availability — measured in percentages Important The Availability section also displays the listener’s availability during the specified time range. • Connection Time — measured in milliseconds • Response Time — measured in milliseconds Managing Oracle Database Systems: User and Reference Guide 125 Chapter 2—Using Foglight for Oracle To investigate a single instance, select the requested instance from the list at the top of the panel. The display switches to instance usability. For details, see “Reviewing Instance Usability” on page 123. Reviewing Locks Activity at the Cluster Level The Activity Locks panel for a RAC comprises the following panes: • Locks Activity • Lock Tree (Current) • Locks Historical Tree Locks Activity The Locks Activity pane displays a breakdown of all lock types and their general properties. Name Description Lock Name The name of the lock. Gets The total number of enqueue requests or enqueue conversions for this type of enqueue. Gets Rate The rate of lock gets. Gets (percent) The percentage of lock gets for this type of lock within the total number of locks. Enqueue Waits The total number of times an enqueue request or conversion resulted in a wait event. Waits Rate The number of wait events per second. Wait (percent) The percentage of wait events for this type of lock within the total number of locks. Managing Oracle Database Systems: User and Reference Guide 126 Chapter 2—Using Foglight for Oracle Name Description Lock Wait The total amount of time spent waiting for the enqueue or enqueue conversion. Average Lock Duration The average length of time for a lock, in milliseconds. Lock Efficiency The percentage of lock requests that did not escalate to a blocking lock. Alarms The number of warning, critical, and fatal alarms that were invoked for this Oracle database instance. Holding the cursor over each alarm type displays a dwell that shows the most recent alarms invoked for this database instance, sorted by severity. Click each alarm type to drill down and manage the alarms. See the Foglight Online Help, Monitoring System-Wide Alarms for details on the alarm information. Lock Tree (Current) The Lock Tree (Current) pane displays current data, which is available only in real-time mode. Clicking any locking or locked session table row displays a popup that is used for viewing further details. For information see “Session Details Popup” on page 127. Name Description PDB In multitenant architectures, the name of the pluggable database where the lock took place. Session The identifier for a session that holds or acquires the lock. Serial The session’s serial number. The combination of SID and serial number provides a unique identification of the session. DB User The database user name. Object Wanted The object requested for lock. Lock Type The type of user or system lock. The locks on the user types are obtained by user applications. Any process that is blocking others is likely to be holding one of these locks. The user type locks are: • TM — DML enqueue • TX — Transaction enqueue • UL — User supplied Managing Oracle Database Systems: User and Reference Guide 127 Chapter 2—Using Foglight for Oracle Name Description Lock Mode The lock mode in which the session holds the lock. This row can have one of the following values: • None • Null • Row-S (SS) • Row-X (SX) • Share • S/Row-X (SSX) • Exclusive Lock Request Mode The lock mode in which the process requests the lock. This row can have one of the values listed below. • None • Null • Row-S (SS) • Row-X (SX) • Share • S/Row-X (SSX) • Exclusive Lock Duration The duration of the lock in seconds. Cross Instance Defines whether the lock is between different cluster nodes: • YES — for global locks • NO — for single node locks Instance The name of the instance to which the blocking or blocked session belongs. Kill Session Used for terminating the selected session, by means of the Kill Session dialog box that is displayed upon clicking this column. Note Killing a session requires having specific privileges. Please provide an Oracle user with these privileges. Alternatively, select the check box Log-in as SYSDBA to provide the credentials of a SYSDBA user. Session Details Popup Name Description SID The session identifier Serial The session’s serial number. The combination of SID and serial number provides a unique identification of the session. Process The process number DB User The database user name Managing Oracle Database Systems: User and Reference Guide 128 Chapter 2—Using Foglight for Oracle Name Description OS The operating system name Program The name of the program that is processing the SQL statements Client The Client computer identifier Host The host computer identifier Current SQL The SQL text of the SQL statement that is currently being processed in this session Locks Historical Tree The Locks Historical Tree pane displays the overall locks activity that took place within the selected time range, as well as locks that exceeded a predefined threshold. This pane comprises the following sections: • Lock Wait Events — a chart that displays all of the lock wait events, plotted over time. This chart can be viewed in either of the following views: • Baseline — displaying only lock wait events. • Breakdown — displaying both lock and non-lock wait events. • Locks Exceeding a predefined threshold — displays locks that reached or exceeded the minimal duration, in seconds, which a lock should reach or exceed in order to be collected. Note The minimal duration’s length is defined using the Locks pane of the Collection Settings view. Important When Foglight Performance Analysis is installed and configured, the minimal lock duration limitation does not exist. Therefore, in instances that collect data using Foglight Performance Analysis, the section is called Locks, and the option of defining a minimal duration for a lock is disabled in the Databases Administration dashboard. Reviewing Locks Activity at the Instance Level The Activity Locks panel, which displays lock summary information for an Oracle instance, comprises the following panes: • Locks Wait Events chart The content of this pane is similar to that of the chart displayed under the Activity > Locks > Locks Historical Tree pane at the Whole Cluster mode. For more information, see “Locks Historical Tree” on page 128. • Locks Activity pane — identical to the Locks Activity pane in the Whole Cluster mode. For more information, see “Locks Activity” on page 125. • Locks Tree (Current) pane — identical to the Locks Tree (Current) pane in the Whole Cluster mode. For more information, see “Lock Tree (Current)” on page 126. Managing Oracle Database Systems: User and Reference Guide 129 Chapter 2—Using Foglight for Oracle • Locks Historical Tree — see “Locks Historical Tree” on page 129. Locks Historical Tree The Locks Historical Tree pane displays the locks activity and the locks tree summary, plotted over time. This pane includes the sections listed below. • Locks Activity — this section, which is identical to the Locks Activity section in the Locks Overview pane, displays the lock wait (average active sessions), which can be plotted as either: • Baseline — lock wait time plotted over time. • Breakdown — lock wait and non-lock wait plotted over time. • Locks Exceeding a predefined threshold — displays locks that reached or exceeded the minimal duration, in seconds, which a lock should reach or exceed in order to be collected. The minimal duration’s length is defined using the Locks pane of the Collection Settings view. For details, see section “Locks” on page 188. Important When Foglight Performance Analysis is installed and configured, the minimal lock duration limitation does not exist. Therefore, in instances that collect data using Foglight Performance Analysis, the section is called Locks, and the option of defining a minimal duration for a lock is disabled in the Databases Administration dashboard. Viewing Detailed Sessions Data Use the Sessions panel to track the activity of all currently running sessions, as well as their resource consumption. Note The refresh rate of the data displayed on the Sessions panel can be set by selecting a value from the Refresh interval drop-down list, which appears on the panel’s upper right side. Important The Sessions panel provides several significant features, which highly enhance data retrieval and display capabilities, when Performance Analysis is installed and configured. This panel can also be used to carry out the tasks described in the following sections: • Viewing detailed resource consumption • “Viewing the host CPU usage” on page 130 • “Tracking the number of logon attempts per second” on page 130 • “Viewing the instance workload by sessions” on page 130 Managing Oracle Database Systems: User and Reference Guide 130 Chapter 2—Using Foglight for Oracle • “Monitoring the active/inactive sessions ratio” on page 131 • “Monitoring the foreground/background sessions ratio” on page 131 • “Viewing the workload incurred by a single session’s activity” on page 131 Viewing detailed resource consumption Use the Resource Breakdown section to view the consumption of various resources as a percentage of the combined workload of all sessions during the specified time range. Note The consumption of the CPU Usage and CPU Wait resources is displayed only when Performance Analysis is installed and configured. In addition to the resource table, from this section it is possible to view more in-depth data, as follows: • Viewing active wait events for a specific resource — by clicking a specific resource within the table. The Active Wait Events popup appears, with the clicked resource selected in the Resource drop-down list. • Viewing all wait events — by clicking the All Wait Events link at the lower right corner of the section. The Active Wait Events popup appears, with the Resource drop-down list unfiltered and displaying the entire list of wait events. Viewing the host CPU usage The Host CPU Usage section contains the CPU Usage spinner, which provides a representation of the total percentage of CPU resources being consumed on the monitored host by all of the operating system’s processes, both Oracle and non-Oracle. This component spins faster as CPU consumption increases. The CPU usage spinner changes its color when CPU overload is detected, as such an overload significantly degrades the system’s performance. Tracking the number of logon attempts per second The Logons section contains a chart, whose contents change depending on whether Performance Analysis is installed and configured: • Without Performance Analysis, this chart displays the number of logon attempts, plotted over the specified time range • If Performance Analysis is installed and configured, this chart displays the number of successful logons, plotted over the specified time range An unexpected, continuous increase of the number of logon attempts, which constantly exceed their baseline, can indicate a need to review the resource consumption. Viewing the instance workload by sessions The Workload (Average Active Sessions) section contains a chart that displays the activity (system workload), plotted over the specified time range. Use this section to toggle between the following views: Managing Oracle Database Systems: User and Reference Guide 131 Chapter 2—Using Foglight for Oracle • Workload — displays the general workload (average active sessions) during the specified time range. • Breakdown — displays the workload, distributed by the various wait event categories. The colors of the categories match the colors of the various resources displayed in the Resource Breakdown section. Monitoring the active/inactive sessions ratio The Active/Inactive Sessions section displays the percentage of active sessions within the entire number of Oracle sessions that ran during the specified time range. Active sessions are non-system sessions that are actively processing in Oracle, or are waiting on locks (blocked). An unexpected increase of the number of active sessions can indicate one of the following situations: • A massive workload — the system handles more users than it was initially designed to do • A bottleneck — lack of system resources prevents users from carrying out their transactions, resulting in wait events and an increasing number of sessions that remain active for prolonged periods Monitoring the foreground/background sessions ratio The Foreground/Background Sessions section displays all of the sessions that ran during the specified time range, distributed as follows: • Background processes — these are mostly user connections • Foreground processes — these are usually Oracle-related processes The background processes’ level of activity should be stable. An unexpected increase in this metric’s value can indicate system overload. An unexpected increase in the foreground processes’ level of activity can indicate that the system handles more users than it was initially designed to do. Viewing the workload incurred by a single session’s activity The Sessions List section contains a table that lists all currently running Oracle sessions, displaying each session in a separate row. Important The Sessions table provides enhanced capabilities, which are available only when Performance Analysis is installed and configured, such as display by resource and Top Sessions view. Note If Foglight for Oracle is monitoring a multitenant architecture, the Sessions table displays the PDB column, which indicates on which pluggable database the session is running. Unlike other panels in the Activity drilldown, which provide summarized information for all sessions, the Sessions table provides a breakdown view of a single session’s activity, and can thereby be used for carrying out the following tasks: • Sorting the table — in order to find the single most resource-intensive session in terms of CPU consumption, current wait time, amount of I/O performed, and so on. Managing Oracle Database Systems: User and Reference Guide 132 Chapter 2—Using Foglight for Oracle • Viewing detailed information about the session — clicking a number on the table’s SID column displays the Session Details panel, which provides detailed information about the selected session in several panes. For details, see “Viewing the Session Details page” on page 132. Viewing the Session Details page The various sections of the Session Details page provide detailed information about the selected session. The following sections are displayed in the Session Details page only if Foglight Performance Analysis is installed and configured: • Resource Breakdown — the consumption of various resources as a percentage of the combined workload of all sessions during the specified time range • Activity — distribution by active and inactive sessions • Usage — memory and CPU utilization plotted over the specified time range • All of the panes of the Session Summary table, except Session Overview The screen components described in the following sections are available also if Performance Analysis is not configured: • Session Details • “System process identification” on page 132 • “Session Details table” on page 133 Oracle Session Identification The Session Details section displays several parameters that provide general information about the selected session, as follows: Parameter Description Session ID The session unique identifier (Session Process ID). The unique number that Oracle has assigned to identify the selected session. Logon Time The time when the user logged on to the session. Status Indicates whether the session is active or inactive. DB User The Oracle login name for this session. Program The name of the program the user is running to access Oracle (in background sessions, displays the name of the process or thread that runs the session). Waiting for: The resource on which the session is waiting. System process identification The System Process Details section displays the following fields: • OS User — the name of the account with which the user logs in to the operating system Managing Oracle Database Systems: User and Reference Guide 133 Chapter 2—Using Foglight for Oracle • Machine — the name of the host from which the session is running Session Details table The Session Details table contains several panes, of which only Session Overview is available when Performance Analysis is not installed and configured. With Performance Analysis, Session Overview contains a chart that plots the session’s SQL executions over the specified time range, and a table that displays the resource utilization, divided by metrics, over this period. If Performance Analysis is not installed and configured, the Session Overview pane displays, instead of a chart, the text of the currently running SQL statement. The panes available only when Performance Analysis is installed and configured are: • Session Statistics — contains a table that displays metrics pertaining to the following classes: Cache, General, I/O, Lock, Redo, SQL, and User. • Open Cursors — contains a table that lists all cursors that were open during the selected time range, their SQL text, CPU usage, active time and the various wait events they experienced • SQL Summary — displays summarized data about each of the SQL Statements that were executed by the session • Session Locks — displays all locks held or requested by the selected session • Datafile I/O — displays the time each datafile spent on I/O wait events during the selected time range Viewing Datafiles I/O Activity at the Cluster Level The Overall Datafiles I/O panel displays the I/O activity of all datafiles on the cluster. This panel includes the following sections: • Total Physical I/O chart • Datafiles I/O Summary table Physical I/O chart The Physical I/O chart displays the physical I/O activity within the cluster’s datafiles, in either of the following methods: • Baseline — displays the physical I/O activity along with the allowed range of typical observed behavior. Use the chart, in its baseline form, to display either a specific metric or all of the metrics. Metric values and typical behavior are plotted over time. • Breakdown — provides a color-coded display of the physical activity of all instances within the RAC. Select whether to display the total physical I/O, only physical reads or only physical writes. Click the arrow to the right of the Physical I/O title to display data for the total physical activity, or only for physical reads or writes. Managing Oracle Database Systems: User and Reference Guide 134 Chapter 2—Using Foglight for Oracle Datafiles I/O Summary Table The Datafiles I/O Summary table displays all of the I/O activity details of all datafiles on the cluster. Clicking the datafile’s name displays the Physical I/O Activity page, which presents the datafile’s physical read and writes, plotted over the time range, along with the overall physical reads and writes. Important To view the specific datafile’s storage data, click the link Datafile Storage, on the upper right side of the Physical I/O Activity page. For details, see “Datafile-specific Page” on page 144. Viewing Datafiles I/O Activity at the Instance Level The Overall Datafiles I/O Activity panel displays the I/O activity of all datafiles on the instance. This panel includes the following sections: • Total Physical I/O Chart • Datafiles I/O Summary Table Total Physical I/O Chart The Total physical I/O chart provides a graphic representation of the physical I/O activity within the instance’s datafiles. This representation is displayed in either of the following methods: • Baseline — displays the total physical I/O activity, including both physical reads and writes. • Breakdown — displays the physical activity, divided by physical reads and physical writes. Datafiles I/O Summary Table The Datafiles I/O Summary table displays all of the I/O activity details of all datafiles on the instance. Clicking the datafile’s name displays the Physical I/O Activity page, which presents the datafile’s physical read and writes, plotted over the selected time range, along with the instance physical reads and writes. In multitenant architectures, the PDB column identifies the pluggable databases that use the datafiles. Important To view the specific datafile’s storage data, click the link Datafile Storage, on the upper right side of the Physical I/O Activity page. For details, see “Datafile-specific Page” on page 144. Viewing User-Defined Collections Use the User-defined Collections panel to see customized collections, which were created using the User-defined Collections view in the Databases Administration dashboard, either during the most recent sample (Last Sample) or plotted over the specified time range (Selected Period). Important The User-defined Collections panel is used only for displaying the user-defined collections; any creation or management operation of these collections is carried out using the User-defined Collections view in the Databases Administration dashboard, at the whole cluster or single instance levels only (not from within a RAC node). For details, see “Configuring User-defined Collections” on page 194. This panel comprises the following sections: Managing Oracle Database Systems: User and Reference Guide 135 Chapter 2—Using Foglight for Oracle • The Collections column, on the left of the panel — displays all of the names of the existing user-defined collections. • SQL text grid — displays the text of the user-defined SQL query. By default, this grid displays the query’s short text. To display the query’s full text, click View full text (a toggle). • The collection details table — displays all of the fields contained in the query. Each of the table’s columns shows the field’s display name, while each row represents the records taken at each sample. Pluggable Databases Drilldown The Pluggable Databases drilldown contains the Pluggable Databases pane, which displays a table of all pluggable databases (PDB), as well as the container databases (CDB), which are monitored in the instance. For details, see “Pluggable Databases List table” on page 135. Selecting a row in the table displays the details of the selected container in the following panes: • Overview — shows general details about the selected database, such as name, availability, and workload. For details, see “Overview tab” on page 135. • Activity within Instance/RAC — shows the activity of the instance as part of the overall instance or RAC activity. For details, see “Activity within Instance/Cluster tab” on page 136. Pluggable Databases List table The Pluggable Databases List table includes the following columns: • Sev — indicates the pluggable database maximum severity level • Name — the pluggable database name • ID — the ID number of the pluggable database • Up Since — displays the date and time when the pluggable database was last started • Workload (% of Instance) — the percentage the pluggable database’s workload constitutes within the entire instance/RAC • CPU (% of Instance) — the percentage the pluggable database’s CPU consumption constitutes within the entire instance/RAC • I/O (% of Instance) — the percentage the pluggable database’s I/O activity constitutes within the entire instance/RAC • Shares — the portion of the system's CPU resources that is allocated to the selected pluggable database Overview tab The Overview tab contains the following sections, which display general information about the pluggable database selected in the Pluggable Databases List table: • Name — the name of the pluggable database Managing Oracle Database Systems: User and Reference Guide 136 Chapter 2—Using Foglight for Oracle • Open mode — the pluggable database’s state; the possible values are: mount, read-write, read-only, and migrate • Created on — the day on which the pluggable database was created • Cloned from — if this pluggable database was cloned, the name of the originating pluggable database • Availability — displays the selected pluggable database’s availability within the selected time range • Storage — displays the pluggable database's storage compared with the storage of the entire instance or RAC • Workload — displays the pluggable database’s workload during the specified time range • Workload Across Cluster — if the selected pluggable database is part of a RAC, displays the balance between the various pluggable databases within the RAC Activity within Instance/Cluster tab The Activity within Instance tab contains the following sections, which display information about the pluggable database selected in the Pluggable Databases List table: • Workload-related Metrics — displays a list of metrics for the pluggable database • Workload — displays the pluggable database’s workload compared with the workload of the entire instance or RAC • Selected metric — displays the activity of the metric selected in the Metric table, compared with the workload of the entire instance or RAC Storage Drilldown The Storage drilldown comprises the panels described in the following sections: • “Tablespaces and Datafiles Panel” on page 137 • Redo Logs — for details, see “Redo Logs Panel” on page 146. • Archive Logs — for details, see “Archive Logs Panel” on page 148. Note The layout of the panels mentioned above is common for both a RAC and database instance. • Invalid Objects — for details, see “Invalid Objects Panel” on page 149. • Database Backup — for details, see “Database Backup Panel” on page 150. Managing Oracle Database Systems: User and Reference Guide 137 Chapter 2—Using Foglight for Oracle Tablespaces and Datafiles Panel Click this link to access the Datafiles page The Tablespaces page Click here to access a specific tablespace The Datafiles page Click any of these rows to access a datafile-specific page Click this link to access the Tablespaces page Managing Oracle Database Systems: User and Reference Guide 138 Chapter 2—Using Foglight for Oracle The Tablespaces and Datafiles panel comprises the sections described in the following sections: • Tablespaces page — displayed by default. When switching to other sections of this panel, accessible by clicking the link View all tablespaces on the upper right side of the screen. For details, see “Tablespaces page” on page 138. • Datafiles grid — accessed by clicking the link View all datafiles on the upper right side of the screen. For details, see “Datafiles Grid” on page 140. • Tablespace-Specific page — accessed by clicking the name of the requested tablespace on the Tablespaces Summary table. For details, see “Tablespace-specific page” on page 140. • Datafile-Specific page — accessed in either of the following methods: • Clicking the name of the requested datafile from the Tablespace Details > Datafiles pane, under the tablespace-specific page (see “Tablespace Details Section” on page 143). • Clicking the name of the requested datafile from the Datafiles grid. For details, see “Datafile-specific Page” on page 144. Tablespaces page The Tablespaces page displays the database individual tablespace information, sorted by either percent of space usage or size. Important By default, all types of tablespaces are monitored. To exclude specific types of tablespaces from monitoring, see “Tablespace Settings” on page 189. The Tablespaces page contains the following components: • Top 10 Tablespaces chart — a representation of the sizes of the tablespaces that have the highest used value, displayed according to the selected sort method. For details, see “Top 10 Tablespaces Chart” on page 138. • Tablespaces Summary table — a full list of the tablespaces, with general parameters and individual alarms. For details, see “Tablespaces Summary Table” on page 139. Top 10 Tablespaces Chart The Top 10 Tablespaces bar chart displays the tablespaces that have the highest used value, sorted by: • Size • Percentage of space usage Name Description Show top Use the Show top options to select which metric the vertical axis displays for the top 10 tablespaces: • Size — size in megabytes. • % Space Usage — percent. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Name Description Percent (chart vertical axis) The percentage of the database tablespace pages currently being utilized. Megabytes (chart vertical axis) The database tablespace utilized size in megabytes. Tablespaces (chart horizontal axis) The names of the top 10 most utilized tablespaces. Tablespace bar The bar chart representation of the utilized tablespaces. Click a specific tablespace representation to drill down further to the Tablespace-specific page. For details, see “Tablespace-specific page” on page 140. Tablespaces Summary Table The Tablespaces Summary table shows the status for all of the database tablespaces. To view details on a specific tablespace, click the tablespace. Name Description PDB In multitenant architectures, identifies the pluggable database that uses the tablespace shown in the Tablespace column. Tablespace The name of the tablespace. Click the tablespace name to drill down further to the Tablespacespecific page. For details, see “Tablespace-specific page” on page 140. Total Size The total amount of allocated space (in megabytes) in this tablespace. % Space Usage The percentage of used allocated space in this tablespace. Contents The type of tablespaces content: • PERMANENT • TEMPORARY • UNDO Number of Datafiles The number of datafiles in the selected tablespace. To view a list of all datafiles of which the tablespace consists, and possibly drill-down to a datafile-specific page, click the tablespace name to access the Tablespace Details > Datafiles pane. Estimated Until Full Estimated number of days until tablespace is full. If the tablespace used space did not grow during the selected period, this field shows the indication no growth. 139 Managing Oracle Database Systems: User and Reference Guide 140 Chapter 2—Using Foglight for Oracle Name Description Autoextensible Datafiles The sum of the Autoextensible OS Datafiles and Autoextensible ASM Datafiles metrics. Alarms The number of warning, critical, and fatal alarms for this Oracle database tablespace. For further details, see “Alarms” on page 52. Datafiles Grid The Datafiles grid displays a list, sorted by usage, of all datafiles used by the database. Name Description PDB In multitenant architectures, identifies the pluggable database that uses the datafile shown in the File Name column. File Name The file name of the datafile, including the full directory path. Click the file name to drill down further to the Datafile-specific panel. For details, see “Datafile-specific Page” on page 144. Tablespace The name of the tablespace that contains the selected datafile. Size The current size of the datafile in megabytes. Used The current used megabytes associated to the datafile, as a percentage of datafile’s allocated size. Autoextend Indicates whether the datafile is auto-extensible. Tablespace-specific page The Tablespace-specific page charts used and free space, as well as the fragmentation level of the tablespace (to determine how much space can yet be utilized). The datafiles that make up the individual tablespace appear on the bottom pane of the dashboard. This page comprises the following sections: • Tablespace Properties section — for details, see “Tablespace Properties Section” on page 141. • Tablespace Properties alarms — for details, see “Tablespace Properties Alarms” on page 141. • Tablespace Properties table — for details, see “Tablespace Properties Table” on page 141. • Space Utilization chart — for details, see “Space Utilization chart” on page 142. • Free Space Fragmentation chart — for details, see “Free Space Fragmentation chart” on page 143. • Tablespace Details section — for details, see “Tablespace Details Section” on page 143. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle • Datafiles table — for details, see “Datafiles Table” on page 143. • All Metrics table — for details, see “All Metrics Table” on page 143. Tablespace Properties Section The tablespace properties are provided using the following panes: • Tablespace properties alarms • Tablespace properties table Tablespace Properties Alarms Name Description Alarms The number of warning, critical, and fatal alarms for the selected tablespace. For further details, see “Alarms” on page 52. Tablespace Properties Table Name Description Status Indicates the current status of the tablespace — Online or Offline. PDB In multitenant architectures, shows the pluggable database that was listed beside the selected tablespace in the Tablespaces Summary table. Extent Management Indicates the method used for managing extents in this tablespace: • DICTIONARY — within the Oracle data dictionary • LOCAL — within the tablespace itself 141 Managing Oracle Database Systems: User and Reference Guide 142 Chapter 2—Using Foglight for Oracle Name Description Allocation Type The type of extent allocation to be used for this tablespace. • USER — allocates extents as per the storage clause provided with the object or defaulted at tablespace level. • UNIFORM — allocates the uniform sizes for the extents. • SYSTEM — automatically allocates system-generated extent sizes. Force Logging Indicates if the tablespace is in Force Logging mode, which logs all changes to all objects in the tablespace, except changes to temporary segments. The possible values are Yes and No. Logging Indicates if a redo log is to be generated for the tablespace. The possible values are Logging and No. Segment Space Management Indicates how free and used segments are managed. • MANUAL — using free lists • AUTOMATIC — using bitmaps Default Table Compression Indicates the status of default table compression; either Enabled or Disabled. Retention Indicates if unexpired undo blocks are retained in the tablespace. • NOT APPLY — used for tablespaces other than the undo tablespace. • GUARANTEE — applies only to undo tablespaces. • NOGUARANTEE — applies only to undo tablespaces. Big File Indicates if the tablespace is a BIGFILE tablespace (Yes) or a small file tablespace (No). BIGFILE tablespaces are supported only for locally managed tablespaces with automatic segment-space management. Block Size The size of the memory block used by this tablespace in megabytes. Encrypted Indicates whether the tablespace data is encrypted (Yes) or not (No). Space Utilization chart Name Description Total Total amount of available memory for this tablespace, plotted over time. Used Amount of available memory that is used for this tablespace, plotted over time. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Free Space Fragmentation chart The Free Space Fragmentation chart displays the results of the fragmentation analysis, whose purpose is to map the fragments by the extent to which they are usable, as shown below. Important As fragmentation analysis is highly resource-consuming, it is advisable not to carry out this operation in operation environments that experience heavy resource contention. Name Description Fully Usable The total sum of fragments that are bigger than, or equal to, the maximum extent in the tablespace. Partly Usable The total sum of fragments that are smaller than the maximum extent in the tablespace, but bigger than, or equal to, the minimum extent in the tablespace. Unusable The total sum of fragments that are smaller than the minimum extent in the tablespace. Tablespace Details Section The tablespace details are provided using the following panes: • Datafiles table • All Metrics table — for details, see “All Metrics Table” on page 143. Datafiles Table Name Description Datafile name The datafile name, including the full directory path. Click the datafile name to drill down further to the Datafile-specific panel. For details, see “Datafile-specific Page” on page 144. Size The current size of the datafile in megabytes. Filesystem Free Available space on the file system as a percentage of file system size, rounded up to one decimal. If the filesystem resides on an Automatic Storage Management (ASM) disk, retrieving this information requires having the ASM instance installed and configured. Alarms The number of warning, critical, and fatal alarms for this Oracle database tablespace. All Metrics Table This table provides the following information for tablespace metrics. • Metric — the metric name; can have one of the values listed in the table below. 143 Managing Oracle Database Systems: User and Reference Guide 144 Chapter 2—Using Foglight for Oracle • Value — current value Metric Name Description Total The total amount of space in the selected tablespace. Free Space The total amount of free space in the selected tablespace. Used The amount of space used in the selected tablespace. Allocated Space Used Percentage The percentage of used space for non auto-extensible files in the selected tablespace. Free Space Percentage The percentage of free space. Minimum Fragment Size The size of the smallest fragment in the selected tablespace. Maximum Fragment Size The size of the largest fragment in the selected tablespace. Average Fragment Size The average size of fragments in the selected tablespace. Recycle Free (MB) The total amount of space used by the recycle bin within the allocated free space of the tablespace. Recycle Free (%) The percentage of recycle bin space which occupies the allocated free space of the tablespace. Estimated Until Full Estimated number of days until tablespace is full. If the tablespace used space did not grow during the selected period, this field shows the indication no growth. Autoextensible OS Datafiles Number of autoextensible OS datafiles (DF). Autoextensible ASM Datafiles If storage is managed using Automatic Storage Management (ASM), and the ASM instance is installed and configured, displays the number of autoextensible ASM datafiles (DF). Manual OS Datafiles Number of manual OS datafiles (DF). Manual ASM Datafiles If storage is managed using Automatic Storage Management (ASM), and the ASM instance is installed and configured, displays the number of manual ASM datafiles (DF). Datafile-specific Page The individual Datafile properties panel displays general properties of the selected datafile, as well as a representation of the used and free space that make up the specific datafile. The Oracle database datafile-specific panel comprises the sections listed below. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle • Datafile Properties • Space Usage — for details, see “Space Usage” on page 145. • Fragmentation — for details, see “Fragmentation” on page 146. Datafile Properties Name Description Alarms The number of warning, critical, and fatal alarms for the selected datafile. Property Label Property Description Datafile The name of the datafile, including the full directory path. Tablespace The name of the tablespace to which the datafile belongs. PDB In multitenant architectures, the name of the pluggable database. Size The size of the datafile in megabytes. Used The used size of the datafile in megabytes. % Used The percentage of the datafile used. Status Indicates whether the datafile is available: • AVAILABLE • INVALID Autoextend Indicates whether the datafile is auto-extensible. Space Usage Details are provided for: 145 Managing Oracle Database Systems: User and Reference Guide 146 Chapter 2—Using Foglight for Oracle • File Space Usage • Filesystem Space Usage Note For datafiles managed by ASM, the filesystem space usage cylinder is available only if the ASM instance is installed and configured. For autoextensible datafiles, the file space usage cylinder is not available. Name Description Used (percent) The percentage of the datafile being used. Used (megabytes) The number of megabytes being used. Total (megabytes) The total number of megabytes available. Fragmentation Use the Fragmentation section to view the file fragmentation status of used and free extents, using a fragmentation analysis triggered by clicking Analyze Fragmentation. Caution Fragmentation analysis is a resource-consuming operation. Carrying out this operation is strongly discouraged in production environment with heavy resource contention. Name Description Fully Usable The amount of memory that is not being used by a datafile and is viable; represented by the blue portion of the horizontal memory bar. This bar displays the file fragmentation snapshot of used and free extents. Partly Usable The amount of datafile memory that is partly usable; represented by the violet portion of the horizontal memory bar. Unusable The amount of tablespace memory that is unusable; represented by the pink portion of the horizontal memory bar. Used The amount of tablespace memory that is used; represented by the orange portion of the horizontal memory bar. Redo Logs Panel The Redo Logs panel, which displays details regarding all redo logs, either for the Oracle instance or for all nodes of an Oracle RAC, comprises the following sections: • Redo Logs Storage Groups table — click on a group list row to see the list of members in that group. • Members of Group [n] table — for details, see “Members of Group [n] Table” on page 147. Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Redo Logs Storage Groups Table The redo log records all changes made in datafiles. Name Description Group # The redo log group number. Group Status The status of the redo logs indicates which logs are in use and whether there are any odd status codes, such as stale log indications or indications of corrupt redo logs. The log files can have one of the following status values: • USED — indicates either that a log had just been added (never used), or that a RESETLOGS command has been issued. • CURRENT — indicates a valid log that is in use. • ACTIVE — indicates a valid log file that is not currently in use. • CLEARING — indicates a log is being re-created as an empty log due to database administrator action. • CLEARING CURRENT — indicates that a current log is being cleared of a closed thread. If a log stays in this status, it could indicate there is some failure in the log switch. • INACTIVE — indicates that the log is no longer needed for instance recovery but may be needed for media recovery. Archived Indicates if the redo logs are being archived. Sequence # The number assigned by Oracle to each redo log file whenever a log switch occurs and LGWR begins writing to the log file. When redo log files are archived, the archived log retains its log sequence number. Size The size of the log in megabytes. First Change # A system-allocated number (SCN), which identifies the first time a change was recorded in this log. First Time The time when the first SCN in the log was recorded. Alarms The number of warning, critical, and fatal alarms for the selected Oracle database tablespace. Members of Group [n] Table Name Description Member Name The redo log group member name. 147 Managing Oracle Database Systems: User and Reference Guide 148 Chapter 2—Using Foglight for Oracle Name Description Member Status The redo log group member status. This status can have one of the following values • INVALID — indicates that Oracle cannot access the redo log file. • STALE — indicates a log file that is either corrupt or incomplete. If the log file’s group becomes the active group, the file becomes valid again. • DELETED — indicates an online redo log file that has been dropped from the database and deleted from the disk. • IN USE — identifies the currently active redo log file. Archive Logs Panel The Oracle database archive logs panel comprises the following sections: • Archive Log Destinations table • Archive Writes chart — displays the amount (in megabytes) of disk space being used per second for the archive log. Note If the instance is configured not to archive, the Archive Logs panel is blank, displaying only the message The database is not archiving. Archive Log Destinations Table When monitoring a RAC, the contents of the Archive Log Destinations table vary between Whole Cluster and Instance/RAC node view, as in whole cluster view the table also includes the column Instance Name, as described below. Name Description Instance Name Indicates which of the RAC nodes uses the specified destination archive file. Note The Instance Name column is displayed only in whole cluster view. From the Archive Destination column and forwards, the two tables are identical. Archive Destination The name of the destination archive file. Destination Directory The name of the destination directory. Managing Oracle Database Systems: User and Reference Guide 149 Chapter 2—Using Foglight for Oracle Name Description Destination Status The current status of the destination in v$archive_dest. This destination can have one of the following values: • VALID — destination was initialized and is available • INACTIVE — no destination information exists • DEFERRED — destination was manually disabled by the user • ERROR — an error occurred during open or copy • DISABLED — disabled after error • BAD PARAM — parameter has errors • ALTERNATE — the destination is in an alternate state • FULL — the destination’s quota size has been exceeded Destination Total The total size of the archive destination target. Destination Free Space The total amount of free space available for the archive destination target. Destination Usable Note The total amount of usable space available for the archive destination target. The Destination Total, Destination Free Space, and Destination Usable columns display information only if the archive is managed using ASM and the ASM instance is installed and configured. Time to Failure The amount of time before the archive log destination becomes full. Binding The binding setting, which describes whether successful archival is required: • MANDATORY • OPTIONAL Destination Type The type of archived log destination — either PUBLIC or PRIVATE. Transmit Mode The mode in which data is transmitted across the network: • SYNCHRONOUS • ASYNCHRONOUS, or • PARALLELSYNC Valid Type The Redo log type or types that are valid for the destination: • ONLINE_LOGFILE • STANDBY_LOGFILE • ALL_LOGFILES Invalid Objects Panel The Invalid Objects panel displays all of the database’s schemas and the invalid objects detected in each schema, as well as their distribution per schema. Managing Oracle Database Systems: User and Reference Guide 150 Chapter 2—Using Foglight for Oracle This panel comprises the following sections: • Invalid Objects Table • Invalid Objects Distribution chart — see “Invalid Objects Distribution for the selected schema” on page 150. Invalid Objects Table The Invalid Objects table displays each schema in a separate row, together with invalid objects of the following types: • Functions • Packages • Package bodies • Procedures • Types • Views • Other type objects In addition, the table displays the total number of invalid objects and the alarms invoked as a result of the invalid objects detected in the schema. Important By default, all of the schemas are monitored by the Invalid Objects collection. A schema with no invalid objects is displayed with a zero value in all of its columns. To exclude specific schemas from being monitored, use the Exclude List pane under the Collection Settings view in the Databases Administration dashboard. For details, see “Invalid Objects Settings” on page 190. For each object type (for example, Procedures) whose metric value contains data, that is: a number greater than zero, clicking the metric value displays a popup with a detailed list of all invalid objects for the specific object type in the selected schema. Invalid Objects Distribution for the selected schema The invalid objects distribution chart displays the distribution of the invalid objects for the schema selected in the Invalid Objects table. Database Backup Panel The Database Backup panel displays various details regarding the backup of the Oracle database and datafile. This panel contains the sections listed below. • Last Successful Backup — the exact time and date when the most recent full backup of the datafile and database took place. A warning icon appears to the left of this section’s title if the time that elapsed from the last successful backup date exceeds a predefined threshold (by default, seven days). Important The threshold for the number of days that elapsed since the last successful full backup of a specific database can be configured by modifying the settings of the Days Since Full Backup alarm, in the Threshold section of the Alarms view in the Databases Managing Oracle Database Systems: User and Reference Guide 151 Chapter 2—Using Foglight for Oracle Administration dashboard. For details, see “Modifying alarm threshold values” on page 179. • Configuration Parameters — user-defined parameters that configure the backup operation, such as the device type and the backup jobs parallelism. • Backup Jobs — a list of the backup operations that took place during the selected time range. A backup operation can have one of the following statuses: • • • • Completed Failed Completed with errors Completed with warnings Operations with a Completed status are indicated with a green check mark icon ( ), while a warning icon appears near each operation whose status is other than Completed. Each backup job is identified by both the Start Time column, which indicates whether the backup was carried on a database or a datafile, and the Job ID column. Clicking the row displays a popup with the details of the backup job in the selected row. Reviewing Configuration Settings Use the Configuration drilldown to review the current configuration settings of the Oracle agent, as well as tracking the modifications made to these settings. This drilldown comprises the following panels: • SGA — displays the settings and utilization of the SGA memory, using different display for the RAC and instance levels. For details, see “Reviewing SGA Settings at the RAC Level” on page 151 and “Reviewing SGA Settings at the Instance Level” on page 152. • Initialization parameters — displayed at both the instance and the RAC levels. For details, see “Reviewing the Initialization Parameters” on page 153. Reviewing SGA Settings at the RAC Level In Whole Cluster view, use the Memory drilldown to carry out the following operations: • Reviewing SGA Settings • Reviewing SGA Utilization • Reviewing the SGA’s Breakdown • Reviewing the free space reserved for the various SGA pools Reviewing SGA Settings Use the SGA management section to review the current SGA settings for the entire RAC, using a row that contains the following indicators: • Automatic SGA — indicates whether the SGA memory’s automatic memory management is turned on or off. • Max SGA Size — displays the maximum amount available for the SGA memory. Managing Oracle Database Systems: User and Reference Guide 152 Chapter 2—Using Foglight for Oracle • SGA Utilization — displays the average amount of SGA memory, in megabytes, which was consumed during the selected time range • Memory Target — displays the target amount of SGA memory, in megabytes Reviewing SGA Utilization The SGA Utilization section contains a chart that displays SGA-related information in a breakdown format, which provides a representation of the size the following components occupy within the SGA memory: • ASM Buffer Cache (data is available only if the ASM instance is installed and configured) • Buffer Cache • Java Pool • Large Pool • Shared Pool • Stream Pool Reviewing the SGA’s Breakdown Use the SGA Size section to review the breakdown of the SGA within the RAC, by means of the following components: • A chart — providing a representation of the size each of the RAC’s nodes occupies within the SGA • A table — containing the name of the RAC or the specific node, along with columns that indicate the SGA’s breakdown between the RAC’s nodes, using the following metrics: • • • • Fixed size Variable size Database buffers Redo buffers Reviewing the free space reserved for the various SGA pools The SGA Pool Free Space section displays, both graphically using a chart and in a table format, the free space reserved for the following types of SGA pools for each of the RAC’s nodes: • Java pool • Large pool • Shared pool Reviewing SGA Settings at the Instance Level Reviewing SGA Settings Use the SGA management section to review the current SGA settings for the entire RAC, by means of a row that contains the following indicators: • Automatic SGA — indicates whether the SGA memory’s automatic memory management is turned on or off • Max SGA Size — displays the maximum amount available for the SGA memory. Managing Oracle Database Systems: User and Reference Guide 153 Chapter 2—Using Foglight for Oracle • SGA Utilization — displays the average amount of SGA memory, in megabytes, which was consumed during the selected time range • Memory Target — displays the target amount of SGA memory, in megabytes Reviewing SGA Utilization The SGA Utilization section contains a chart that displays the instance’s current SGA size in either of the following views: • Baseline — total SGA space utilization (current SGA size) • Breakdown — displays the size the following components occupy within the instance’s SGA: • ASM Buffer Cache (data is available only if the ASM instance is installed and configured) • Buffer Cache • Java Pool • Large Pool • Shared Pool • Stream Pool Reviewing the SGA’s Breakdown Use the SGA Size section to review the breakdown of the SGA within the instance, by means of the following components: • A pie chart — providing a representation of the size each component described in the table occupies within the instance’s SGA • A table — shows the SGA’s breakdown between the following components: • • • • Fixed size Variable size Database buffers Redo buffers Reviewing the Free Space Reserved for the Various SGA Pools The Free Space section displays, both graphically using a chart and in a table format, the free space reserved for the following types of SGA pools for each of the RAC’s nodes: • Shared pool • Java pool • Large pool Reviewing the Initialization Parameters The Initialization Parameters panel displays the entire list of initialization parameters, as defined by Oracle. These parameters define the characteristics of each Oracle instance when the instance is started. Upon clicking anywhere on a row that displays a specific initialization parameter, a popup appears, listing the properties of the selected parameter: Managing Oracle Database Systems: User and Reference Guide 154 Chapter 2—Using Foglight for Oracle • The parameter description, taken from Oracle’s documentation (also appears under the Description column). • Link to perform searching for further information. • Change History table — displays the entire history (time and dates) when the Foglight for Oracle agent detected the changes made to this parameter since the instance was created. The Initialization Parameters table varies between instance mode and RAC mode. For details about the RAC mode view, see “Reviewing the Initialization Parameters in RAC Mode” on page 154. In Instance mode, the table displays the follow the columns listed below. Name Description Parameter Parameter name. PDB In multitenant architectures, the name of the pluggable database. PDB is Modifiable Hidden by default. In multitenant architectures, displays the list of pluggable databases that have been modified. Current Value The parameter’s current value. Default Indicates whether the parameter’s current value is its default value, as defined by Oracle. Dynamic Indicates whether the initialization parameter is dynamic, that is, can be modified using the ALTER SESSION, ALTER SYSTEM, or ALTER SYSTEM DEFERRED commands while an instance is running. Modified Indicates whether the parameter’s value has been modified. Deprecated Indicates whether the parameter is deprecated, in which case specifying it in a parameter file will invoke a warning message when the instance starts. Type The parameter type, as defined by Oracle. Description The parameter description, taken from Oracle site. PDB_modifiable Indicates whether the pluggable database can be modified. Note This column is not displayed by default. Reviewing the Initialization Parameters in RAC Mode In RAC mode, the Initialization Parameters panel can be accessed by selecting either the Whole Cluster or a specific instance view. Managing Oracle Database Systems: User and Reference Guide 155 Chapter 2—Using Foglight for Oracle In the Whole Cluster view, instance-specific initialization parameters are displayed with the icon to the left of the parameter name, and the names of the various instances are displayed in the Scope column. The following columns in Initialization Parameters table differ in RAC mode: Name Description Parameter Parameter name. A icon to the left of the name indicates that the parameter applies differently to specific instances. Scope Indicates whether the parameter applies to the entire RAC or varies between specific instances. Current Value The parameter’s current value. This value can be either instancespecific or apply to the entire RAC. Reviewing the Alert and Error Logs The Alerts drilldown provides a view on the alert and error logs received for a selected Oracle agent during the specified time range. This drilldown comprises the following panels: • Error Log — see “Reviewing the Error Log” on page 155 • Alert Log — see “Reviewing the Alert Log” on page 156 Note Both the Error Log and Alert Log panels are displayed only at the instance level. Selecting the Error Log or Alert Log menu item from within a RAC, when the Whole Cluster view is selected, displays a message that prompts the selection of the requested instance. Reviewing the Error Log The Error Log panel shows error messages that were selected to be extracted and displayed from the Foglight agent error logs. The error messages appear with their exact date and time, as well as severity level. Managing Oracle Database Systems: User and Reference Guide 156 Chapter 2—Using Foglight for Oracle Only a small number of the error log messages are displayed in this panel. The messages to be displayed, as well as their severity, are determined within the agent and cannot be modified by the user. The type of agents whose error logs are used for the message display is determined according to the selected view, as listed below. • DB_Oracle — if the selected view is a single instance. • DB_Oracle_RAC_Instance — if the selected view is a RAC node. Reviewing the Alert Log By default, the Alert Log panel displays all of the alert log messages. To display messages under a pre-defined name, category and severity, create or edit patterns for these messages in the Match List, which is found in the Alert Log view in the Databases Administration dashboard, and add the corresponding regular expressions to this list (see “Configuring the Alert Log Panel Display” on page 190). To prevent messages from being displayed on the Alert Log panel, add these messages to the Ignore List. Important Only messages that are explicitly defined in the Ignore List will not be displayed. Messages that were not added to either the Match or Ignore lists appear with the name Other, the type of Oracle alert messages, and the severity Information. Therefore, ensure that messages that do not need to be displayed are added to the Ignore List. The Alert Log panel comprises the panes described in the following sections: • “Alert Log Messages Chart” on page 156 • “Alert Log Messages Table” on page 157 Alert Log Messages Chart The Alert Log Messages chart provides a graphic, color-coded representation of the various alert log messages, broken down by their severity level, as described below. • Informational — displayed on the screen, but does not invoke alarms • Warning Managing Oracle Database Systems: User and Reference Guide 157 Chapter 2—Using Foglight for Oracle • Critical • Fatal The severity assigned for each alert, as well as the minimal severity level (Warning or higher) for which alarms are to be invoked as a result of the alert, are defined using the Alert Log view in the Databases Administration dashboard, accessible from the Agent Settings button. For details, see “Configuring the Alert Log Panel Display” on page 190. Alert Log Messages Table The Alert Log Messages table provides the following details about each alert log: • Timestamp — the exact time and date when the alert log message was created • Severity — can have one of the values listed below • Informational • Warning • Critical • Fatal • Message — the title of the Oracle error message • Dump File — this field is populated by a link only if the error message also refers to a dump file. When the link is clicked, the Dump File popup appears, displaying the first 100 lines of the dump file contents. • Name — the alert log name • Type — the alert log type, such as Oracle errors or Oracle alert messages Note Both Name and Type refer to the parameters defined using the Alert Log view in the Databases Administration dashboard. Reviewing Exadata-related Information Starting from version 5.6.5.330, Foglight for Oracle monitors Exadata database machines, both from the hardware and software aspects, using the Exadata drilldown. The Exadata-related information is presented only in RACs where Exadata is configured. While Exadata appears as one of the drilldowns, clicking Exadata opens an Exadata dashboard. The Exadata dashboard comprises the following drilldowns: • Summary — see “Displaying Exadata-related Summarized Data” on page 157 • Performance — see “Reviewing the Exadata Performance” on page 159 • InfiniBand Topology — see “Reviewing the InfiniBand Topology” on page 160 Displaying Exadata-related Summarized Data The Summary drilldown provides summarized data about the performance and resource utilization of the Exadata cells. Managing Oracle Database Systems: User and Reference Guide 158 Chapter 2—Using Foglight for Oracle This drilldown comprises the following components: • Cell Table — see “Cell table” on page 158. • Performance and Resource Utilization section — see “Performance and resource utilization indicators” on page 159. Cell table The cell table provides the following details about each alert log: • ID — displays the cell’s ID • Name — displays the cell’s name • Status — displays an Orange light if the cell is offline, or a green light if the cell is online • Connect Availability — displays a green light if the cell’s availability is 100%, otherwise displays a red light • Up Since — displays the date and time when the instance was last started • Fans (Working/Total) — displays a green light if the number of working fans equals the total number of fans, otherwise displays a red light • Power Supplies (Working/Total) — displays a green light if the number of working units equals the total number of units, otherwise displays a red light • Temperature — displays a green light if the temperature’s status is Normal, otherwise displays a red light Managing Oracle Database Systems: User and Reference Guide Chapter 2—Using Foglight for Oracle Performance and resource utilization indicators The performance and resource utilization indicators includes the following components. Indicator Description Flashcache Hit Ratio (%) Displays the percentage of block requests resolved from the cell flash cache falls below a predefined threshold. This figure is displayed for the selected cells, as well as for all cells. Block retrieval is carried out significantly faster by accessing the flash cache, as this cache resides on highperformance flash SSD. Low flash cache hit ratio indicates that many blocks are being retrieved from the regular disks, possibly resulting in performance issues. Memory Utilization (%) Displays the memory utilization for the selected cells, as well as for all cells. Overly high memory utilization can result in excessive swapping, which is likely to lead to performance issues. CPU Busy Percent (%) Displays the CPU utilization for the selected cells, as well as for all cells. Overly high CPU utilization can result in excessive swapping, which is likely to lead to performance issues. Cell Performance Displays several key indicators for the cell’s performance, as follows: • Run Queue Length • InfiniBand MB Sent Rate • In fin i band MB Sent Rate Cell Offload Efficiency The number of rows processed by the cells for all the queries divided by the number of rows forwarded by the cells to the DB node instances. Higher values of this metric indicate that the cells made good use of their CPUs and high I/O bandwidth to process high number of rows, extract the relevant ones and only forward those to the DB nodes. Reviewing the Exadata Performance The Performance drilldown includes the following components: • Cell table — see “Cell table” on page 158. • Query efficiency statistics • Celldisks statistics 159 Managing Oracle Database Systems: User and Reference Guide 160 Chapter 2—Using Foglight for Oracle Query efficiency statistics Celldisk response time The Celldisk Response Time (ms) section displays the following metrics on all of the celldisks, as well as on the celldisks selected in the cell table: • Celldisks small writes latency — the average latency, in milliseconds, to write small blocks per request. • Celldisks small reads latency — the average latency, in milliseconds, to read small blocks per request. Flash cache statistics Use the Flash Cache Statistics section to review the load on the flash cache and the number of I/O errors encountered there, by displaying the following metrics on all of the celldisks, as well as on the celldisks selected in the cell table: • Flash cache I/O errors — the number of IO errors that were encountered on the FlashCache • Data read from flashcache (MB) — the total number of megabytes that were read from the flash cache during the specified time range. • Data written to flashcache (MB) — the total number of megabytes that were written to the flash cache during the specified time range. Flash cache usage chart The Flash Cache Usage (MB) chart displays the following metrics on all of the celldisks, as well as on the celldisks selected in the cell table: • Flash cache usage — the total amount of space (in megabytes) used on the flash cache. • Flash cache usage (keep objects) — the total amount of space (in megabytes) used only on the keep objects in the flash cache. Flash cache request misses chart The Flash Cache Request Misses (requests) displays the total number of read IO requests that did not find all data in Flash Cache and therefore had to perform physical read operations. This figure is displayed for all of the celldisks, as well as on the disks selected in the cell table, and is split into two charts: • Total number of request misses of all types. • Total number of request misses resulting from read IO requests for keep objects. Celldisks statistics The Celldisk Statistics section displays the rate (megabytes per second) of physical reads and writes that were carried out during the selected time range on all of the disks, as well as on the disks selected in the cell table. Reviewing the InfiniBand Topology Use the InfiniBand topology drilldown to monitor the InfiniBand switches and nodes. This drilldown includes the following sections: Managing Oracle Database Systems: User and Reference Guide 161 Chapter 2—Using Foglight for Oracle • Switches — includes a table that displays all of the InfiniBand switches, which are used for transferring information from the database to the storage cells and conversely. This section provides the following details: • Number of InfiniBand switches • Subnet manager on switch — identifies which of the switches is a master switch. • Link status — the InfiniBand Switch's current link status, as indicated during the last sample. • Health status — the InfiniBand Switch's current health status, as indicated by the success or failure of the environment test. • Nodes — includes a table that displays all of the InfiniBand nodes. This section provides the following information for each node: • Node name • Node IP address • Node host channel adapter (HCA) • Node status • Node’s ports on each of the InfiniBand switches • Charts — display the rate (Mb/S) of data sent and received through the InfiniBand switch during the specified time range. Foglight for Oracle Agents This chapter provides a general overview of the various agents of Foglight for Oracle. It also describes the properties for each of the agents. Working with Agent Properties When an agent connects to the Foglight Management Server, it is provided with sets of properties that it uses to configure its correct running state. Each agent is provided with a combination of two types of properties: agent properties and shareable properties. While default versions of these properties are installed with Foglight for Oracle, it is possible to edit the default shareable and agent properties, configure agent properties that apply only to a specific agent instance, and create edited clones of shareable properties that are used by a subset of the agents of a certain type. For more information about working with agent properties, see the Foglight Administration and Configuration Guide. Caution The Agent Management and Agent Debug Options sections should not be modified, unless instructed explicitly by Quest Support. Any modification must be made under the instruction and guidance of Quest Support. Managing Oracle Database Systems: User and Reference Guide 162 Chapter 2—Using Foglight for Oracle DB_Oracle Agent Overview The DB_Oracle Agent monitors an Oracle instance by collecting information, using SQL queries (and in certain cases, operating system calls to the computer on which the instance is running). One agent needs to be deployed to each independent Oracle instance within the enterprise that is to be monitored. Oracle instances can be easily detected and deployed for monitoring through the Database Discovery cartridge. For more information, see “Adding and Configuring Agents” on page 14. DB_Oracle Agent Properties The following sets of agent properties are modifiable: • Properties configurable using the Connection Details screen The property sets listed below can be configured using the Connection Details screen. For details, see “Defining Connection Details” on page 166. • Monitored Connection Profile • OS Host Platform • OS Monitored Connection Profile (UNIX) • OS Monitored Connection Profile (Windows) • Properties configurable using the Collection Frequencies screen The property set Collection Details can be configured using the Collection Frequencies screen. For details, see “Defining the Collection Settings” on page 188. • Properties configurable using the Collection Settings screen The following property sets can be configured using the various panes of the Collection Settings screen: • Space Monitoring (Tablespaces) — for details, see “Tablespace Settings” on page 189. • Locks — for details, see “Locks” on page 188. • Sessions — for details, see “Top Sessions” on page 188. • Top SQL Statements — for details, see “Top SQL Statements” on page 188. • Locks History Tree — for details, see “Locks” on page 188. • Invalid Objects — for details, see “Invalid Objects Settings” on page 190. • Properties configurable using the Alert Log screen The property set Alert Log can be configured using the Alert Log screen. For details, see “Configuring the Alert Log Panel Display” on page 190. DB_Oracle_RAC Agent Overview The DB_Oracle_RAC Agent monitors an Oracle RAC by collecting information using SQL queries (and in some cases, operating system calls to the computer on which the RAC is running). One agent needs to be deployed to each Oracle RAC within the enterprise that is to be monitored. Managing Oracle Database Systems: User and Reference Guide 163 Chapter 2—Using Foglight for Oracle When monitoring a RAC database instance, an DB_Oracle_RAC_Instance Agent must be deployed to each instance in the RAC. For more information, see “DB_Oracle_RAC_Instance Agent Overview” on page 163. Oracle RACs can be easily detected and deployed for monitoring using the Database Discovery cartridge. For more information, see “Adding and Configuring Agents” on page 14. DB_Oracle_RAC Agent Properties The following sets of agent properties are modifiable: • Properties configurable using the Collection Frequencies screen The property set Collection Details can be configured using the Collection Frequencies screen. For details, see “Defining the Collection Settings” on page 188. • Properties configurable using the Collection Settings screen The following property sets can be configured using the various panes of the Collection Settings screen: • Space Monitoring (Tablespaces) — for details, see “Tablespace Settings” on page 189. • Locks — for details, see “Locks” on page 188. • Sessions — for details, see “Top Sessions” on page 188. • Top SQL Statements — for details, see “Top SQL Statements” on page 188. • Locks History Tree — for details, see “Locks” on page 188. • Invalid Objects — for details, see “Invalid Objects Settings” on page 190. • Properties configurable using the Alert Log screen The property set Alert Log can be configured using the Alert Log screen. For details, see “Configuring the Alert Log Panel Display” on page 190. DB_Oracle_RAC_Instance Agent Overview The DB_Oracle_RAC_Instance Agent monitors an Oracle RAC instance by collecting information using SQL queries (and in some cases, operating system calls to the computer on which the instance is running). One agent needs to be deployed on each Oracle RAC instance within the enterprise that is to be monitored. Oracle RAC instances can be easily detected and deployed for monitoring using the Database Discovery cartridge. For more information, see “Adding and Configuring Agents” on page 14. DB_Oracle_RAC_Instance Agent Properties The following sets of agent properties are modifiable: • Properties configurable using the Connection Details screen The property sets listed below can be configured using the Connection Details screen. For details, see “Defining Connection Details” on page 166. • Monitored Connection Profile Managing Oracle Database Systems: User and Reference Guide 164 Chapter 2—Using Foglight for Oracle • OS Host Platform • OS Monitored Connection Profile (UNIX) • OS Monitored Connection Profile (Windows) • Properties configurable using the Collection Frequencies screen The property set Collection Details can be configured using the Collection Frequencies screen. For details, see “Defining the Collection Settings” on page 188. DB_Oracle_ASM Agent Overview The DB_Oracle_ASM Agent monitors an ASM instance by collecting information, using SQL queries (and in certain cases, operating system calls to the computer on which the instance is running). ASM instances can be easily detected and deployed for monitoring through the Database Discovery Wizard. For more information, see “Adding ASM Instances” on page 34. DB_Oracle_CRS Agent Overview The DB_Oracle_CRS Agent monitors Oracle Clusterware by collecting information, using CRS commands (invoked by connecting to the OS). One agent needs to be deployed per CRS cluster, regardless of the number of nodes that take part in the cluster. Oracle clusterware can be easily configured for monitoring through the Clusterware monitoring wizard. For more information, see “Adding CRS Agents for Monitoring Clusterware” on page 39. DB_Oracle_Exadata Agent Overview The DB_Oracle_Exadata Agent monitors Exadata environment by collecting information using Shell script queries. The agent monitors Exadata cells along with the corresponding InfiniBand switches. The Exadata environment can be easily configured for monitoring through the Exadata monitoring wizard. For more information, see “Adding Agents for Monitoring Exadata Environments” on page 42. Managing Foglight for Oracle Agent Settings You use the Databases Administration dashboard to set options for collecting, storing, and displaying data about monitored Oracle instances. Managing Oracle Database Systems: User and Reference Guide 165 Chapter 2—Using Foglight for Oracle Opening the Databases Administration Dashboard You can edit settings for one or more Oracle instances on the Databases > Administration dashboard. Note If you attempt to select instances of more than one type of database, such as an SQL Server database and an Oracle database, an error message is displayed. To open the Databases Administration dashboard: 1 In the navigation panel, under Homes, click Databases. 2 Select the row check boxes beside one or more Oracle instances. 3 Click Agent settings and then click Administration. The Administration dashboard opens, containing settings for all the selected agents. Settings are broken down into categories, which are organized under an Oracle tree. Tip The list of agents you selected can be found by clicking Selected Agents. Reviewing the Administration Settings Use the Databases Administration dashboard to set options for collecting, storing, and displaying data, which apply to all of the currently selected agents. Click a category of settings on the left (for example: Connection Details) to open a view containing related settings on the right. The metrics defined on the Databases Administration dashboard apply to all of the agents that were selected before opening the Administration dashboard. As a result, the same unit of measure and aggregation value for display are enforced for all currently selected agents. To view the full list of selected agents, click Selected Agents button ( ) at the upper right corner of the screen. To change the list of agents to which the metrics apply, exit the Databases Administration dashboard, select the required agents, and re-open the dashboard. If the settings vary between the selected agents (for example: one agent uses the measurement unit kilobyte, while another uses megabyte), the fields that contain non-identical values are displayed as empty and marked with an Inconsistent Values ( ) icon. Managing Oracle Database Systems: User and Reference Guide 166 Chapter 2—Using Foglight for Oracle Changes made to settings should be saved before selecting another category of settings. To save changes made in an Administration dashboard view: 1 In the Database Administration dashboard, select a category of settings from the left-hand panel. 2 Make changes to settings as necessary. 3 Click Save changes at the bottom of the view. If you attempt to exit the view without saving changes, a Warning dialog box prompts you to confirm your action. Defining Connection Details Use the Connection Details category to define global connection settings, which apply to all instances and hosts selected in the view. You can configure Performance Analysis connectivity, enable user-defined collections, and set VMware connection details. Note The following sections instructs how to define the connection settings for monitored Oracle instances. For details about defining connection settings for monitored ASM instances, see Defining the Connection Settings for the Monitored Oracle Instances The Connection Details view contains a table that displays all the agents that were selected before entering the Databases Administration dashboard. To define the connection settings for the requested agents: 1 Select the check boxes to the left of the agents for which uniform credentials are to be set. To cancel the selection, click Select None and select again. 2 Click Set Credentials. The Edit Instance Credentials dialog box opens. Managing Oracle Database Systems: User and Reference Guide 167 Chapter 2—Using Foglight for Oracle 3 Use the Connection Details section to enter the TNS listener port, or accept the default port (1521). 4 Use the Login Credentials section to type the user name and the password used for connecting to the Oracle instance. 5 Select whether to enable OS monitoring, in order to retrieve OS-related data, such as CPU utilization and memory consumption. For instructions about configuring OS monitoring, see “Configuring OS monitoring credentials” on page 169. Note When enabling OS monitoring, ensure that the selected OS user has the privileges required for monitoring the operating system. For details, see the Foglight for Oracle Release Notes. 6 If Foglight Performance Analysis is installed, select whether to edit the Performance Analysis settings. For details, see “Configuring user-defined Performance Analysis connectivity details” on page 170. 7 Select whether to enable and set user-defined collections. For more information, see “Enabling and setting credentials for user-defined collections” on page 170. 8 If the selected agents reside on a VMware virtual host, click Edit VMWare connection details to define their connection details. For more information, see “Defining the VMware Connection Profile” on page 171. 9 Click Validate to proceed to the next stage of validating the instance’s connectivity. Managing Oracle Database Systems: User and Reference Guide 168 Chapter 2—Using Foglight for Oracle Validating Connectivity and Starting to Monitor the Instances After setting the default credentials for the host, these newly created credentials can now be used by the wizard to attempt to log in to the instances. To validate the instances’ connectivity: 1 Click Validate connectivity. The Verifying Connectivity progress bar appears. At the end of this process, any connectivity issues are listed in the Status column of the instance table. When the connection is successful, the Status column displays the status message Validated, which indicates that the instance connected successfully and the specified Oracle user has the required permissions. If the connection failed verification, the Status column displays one of several connectivity status messages. The messages, causes, and appropriate responses are: • • • • • No valid RAC nodes were detected (for a RAC) — modify the RAC nodes’ credentials. Invalid username/password — check the credentials and try again. Process Error — see the code error under the Details column. Wrong Database Credentials — modify the login credentials. Database Connection Failed — the database to which the instance tries to connect is not running or cannot be accessed. Ensure clearing the check box near the database’s name. • Insufficient Privileges — grant the user the privileges required for connecting to the database, by clicking the status Insufficient Privileges. The Grant Database Privileges dialog box opens. Use this dialog box to specify a SYSDBA (System Database Administrator) user with sufficient privileges. Type a SYSDBA user and password, and then click Grant. If the Sysdba credentials entered were incorrect, the column displays the status Wrong Sysdba credentials. Managing Oracle Database Systems: User and Reference Guide 169 Chapter 2—Using Foglight for Oracle After correcting the mistakes that resulted in the connectivity failure, click again Validate connectivity. 2 If the selected agents have Performance Analysis configured, click Detect Performance Analysis to discover the Performance Analysis agents. The Detect Performance Analysis Ports dialog box opens. Use the field inside this box to replace the default port or to type additional ports. Click Detect when completed. Upon successful completion of this process, the Status column of the instance table displays the status Performance Analysis detected. 3 Click Validate connectivity on the status bar. 4 Click Save Changes. The Applying Modified Settings progress bar appears. Configuring OS monitoring credentials To configure the OS monitoring credentials: 1 Use the Host Credentials section to select whether to log in to the monitored host using the existing host connection details or to edit the host connection details. 2 If the option to edit the host connection details is selected, select whether to disable OS monitoring, which is enabled by default. Note When enabling OS monitoring, ensure that the selected OS user has the privileges required for monitoring the operating system. For details, see the Foglight for Oracle Release Notes. 3 Specify the requested authentication type: • Local user — using the same credentials that were used for running the monitoring software (Agent Manager) on the Oracle host. • Windows — the user name (in the domain\user name format) and password, used for logging in to a Windows-based monitored host. • SSH (DSA) — using the SSH private key, generated via the DSA algorithm. • SSH (RSA) — using the SSH private key, generated via the RSA algorithm. • SSH (login credentials) — the user name and password used for logging in to a UNIXbased monitored host. 4 If any option other than Agent Manager local user is selected, enter the required credentials. Note If either of the UNIX authentication types is selected, to run certain commands that require administrative privileges as root, without having to log in as administrator, select the Use sudo check box. Note It is possible to delete or edit existing credentials by clicking Select from stored credentials and entering the User Credentials dialog box. However, it is highly advisable not to make any changes to existing credentials, as such changes will affect all agents that currently use these credentials. Managing Oracle Database Systems: User and Reference Guide 170 Chapter 2—Using Foglight for Oracle Configuring user-defined Performance Analysis connectivity details The Connection Details view in the Databases Administration dashboard provides the following options for configuring the Foglight Performance Analysis connectivity for the selected agents: • Detecting the Performance Analysis port — either by using the default port or by specifying one or more ports. • Setting user-defined details for the Performance Analysis connectivity. To detect the Performance Analysis port: 1 Select the requested agents. 2 Click Detect Performance Analysis. The Detect Performance Analysis dialog box opens. 3 Select whether to accept the default Performance Analysis port (3566) or to replace it, by typing one or more ports separated by comma. 4 Click Detect. Upon successful completion of the detection process, the Status row remains blank and a appears in the PA column. Agents that failed this process display the status Performance Analysis not detected. To configure Performance Analysis manually: 1 Select the requested agents. 2 Click Set credentials. The Edit Instance Credentials dialog box opens. 3 Click Edit Performance Analysis settings. The Edit Performance Analysis settings dialog box opens. Note If the connection to Performance Analysis has been previously configured by clicking Detect Performance Analysis, this dialog box already contains the set ports and has the check box Enable Performance Analysis selected. 4 Select the check box Enable Performance Analysis. 5 Type the requested Performance Analysis port, or several ports separated by comma. 6 Click Set. Important To maintain the connection to Performance Analysis (activate/deactivate/verify accounts and so on), use the Performance Analysis view. For more information, see “Configuring the Connection to Performance Analysis” on page 193. Enabling and setting credentials for user-defined collections Use the user-defined collections feature to add customized collections to all of the currently selected agents, thereby providing for queries not included in Foglight for Oracle. Use the Connection Details view to enable and configure the creation of user-defined collections. To enable and configure the creation of user-defined collections: 1 Select the requested agents. Managing Oracle Database Systems: User and Reference Guide 171 Chapter 2—Using Foglight for Oracle 2 Click Set credentials. The Edit Instance Credentials dialog box opens. 3 Click Edit Credentials for User-defined Collections. The Edit Credentials for User-defined Collections dialog box opens. 4 Select the check box Enable user-defined collections. 5 Select whether to perform the collection using the existing agent credentials or by creating dedicated credentials. 6 If the user-defined collections are to be enabled using dedicated credentials, type the requested user name and password. 7 Select whether to connect as SYSDBA. 8 Click Set. To add user-defined collections, go to the User-defined Collections view in the Databases Administration dashboard. For details, see “Configuring User-defined Collections” on page 194. Defining the VMware Connection Profile If you have Oracle instances that run on virtual hosts, monitoring such instances requires setting a dedicated connection profile, in order to connect to the requested VMware server. Note The VMware connection profile can be defined only if OS monitoring is enabled. Establishing such a connection is necessary in order to retrieve the Virtualization overhead data, that is, the percentage of CPU that is unavailable to this virtual machine because it is being utilized either by other virtual machines or by VMware itself. The Virtualization Overhead indicator is displayed in both the real-time and history summary pages. To edit the VMWare Connection Profile: 1 Select the requested agents. 2 Click Set credentials. 3 In the Edit Credentials dialog box, click Edit VMWare connection details. The Edit VMWare Credentials dialog box opens. Managing Oracle Database Systems: User and Reference Guide 172 Chapter 2—Using Foglight for Oracle 4 Select the check box Enable monitoring the VMWare CPU layer. 5 Enter the details required for monitoring the CPU distribution data, that is: vCenter or ESX host name, port, VMware user, and VMware password. 6 Click OK. Defining the Connection Settings for the Monitored ASM Instances The Connection Details view contains a table, which displays all of the agents that were selected before entering the Databases Administration dashboard. To define the connection settings for the requested ASM agents: 1 Go to the ASM tab. 2 Select the check boxes to the left of the requested ASM agents. 3 Click Agent settings. 4 Click Configure ASM agents. 5 In the Connection Details view, select the check boxes to the left of the ASM agents for which uniform credentials are to be set. To cancel the selection, click Select None and select again 6 Click Set Credentials. The dialog box Edit credentials of ASM instance opens. Managing Oracle Database Systems: User and Reference Guide 173 Chapter 2—Using Foglight for Oracle This dialog box is identical to the one used for configuring ASM instances for monitoring. For details, see step 8 on page 36. Defining Data Collection and Storage Options The Foglight for Oracle agent collects and stores data at all times, even when the browser window that displays the data is not active. Use the Collection Frequencies view in the Databases Administration dashboard to specify: • Which collections are sampled and stored. • The data collection values when sampling is carried out in offline, online, and real-time frequency modes. • The collection frequency. When a user is currently focusing on a screen, the sampling frequency for all of the collections associated with this screen automatically switches to Real-Time. The collection frequency setting determines the sampling frequencies of the other collections, which do not run in Real-Time mode). The available collection frequencies, which can be selected from the field The selected agent's current collection frequency is, are: • Low — all collections are running in Offline mode, regardless of whether a Client is connected • Normal — the collections' running mode (Online/Offline/Real-time) adjusts dynamically to the Client's connection status (disconnected/connected/focusing on a screen). Managing Oracle Database Systems: User and Reference Guide 174 Chapter 2—Using Foglight for Oracle • High — all collections are running in Online mode, regardless of whether a Client is connected. • The Query timeout for on-demand collections. This setting defines the number of seconds that a query for on-demand collections can run before it times out. On-demand collections are collections whose data is retrieved not by predefined time intervals but upon entering a screen or clicking a button. The default setting of this parameter is 60 seconds, but it can be modified by clicking the number that indicates the parameter’s value in the field Query timeout for on-demand collections:. The Collections table contains the columns listed below. Column Description Collection Enabled Defines whether the selected collections are sampled and stored. Collection Name The name of the collection. This list is sorted in alphabetical order. Offline Frequency (Sec) Used for defining the collection interval, in seconds, in offline mode (for example: 300). Offline frequency refers to the longest interval possible for sampling the monitored instance. Online Frequency (Sec) Used for defining the collection interval, in seconds, in online mode (for example: 60). Real-time Frequency (Sec) Used for defining the collection interval, in seconds, in real-time mode (for example: 20). Only one collection can be sampled at real-time frequency at any given moment. Query Timeout (Sec) The amount of time, in seconds, that elapses before the query times out. Important In the Normal collection frequency, when the browser window that displays Foglight for Oracle is active, the collection frequency mode in the active screen (for example: the Current Sessions panel in the Activity drilldown) switches to the fastest frequency possible - once every 20 seconds. To modify the values of a specific collection: 1 Select the collection’s row in the table. 2 Click Edit. The Edit the Collection dialog box opens. Managing Oracle Database Systems: User and Reference Guide 175 Chapter 2—Using Foglight for Oracle 3 Select whether to enable the collection and storage of the selected collection. 4 Set the collection interval, in seconds, in offline frequency mode (if available). 5 Set the collection interval, in seconds, in online frequency mode (if available). 6 Set the collection interval, in seconds, in real-time frequency mode. 7 Set the query timeout, in seconds. 8 Click Set to apply these settings or Cancel to reject them. Customizing Alarms for Foglight for Oracle Rules Many Foglight for Oracle multiple-severity rules trigger alarms. To improve your monitoring experience, you can customize when alarms are triggered and whether they are reported. You can also set up email notifications. This section covers the following topics: • Introducing the Alarms View • Setting and Modifying Alarm Sensitivity Levels • Modifying Alarm Settings • Configuring Email Notifications • Cloning Agent Settings • Reviewing Rule Definitions Introducing the Alarms View The Alarms view enables you to modify global settings and agent-specific settings for alarms. To open the Alarms view: 1 Open the Administration dashboard as described in “Opening the Databases Administration Dashboard” on page 165. 2 Click Alarms. The list of agents that you selected on the Databases dashboard is shown in the upper right corner of the view. Managing Oracle Database Systems: User and Reference Guide 176 Chapter 2—Using Foglight for Oracle 3 From the Alarms view, you can complete the following tasks: • • • • Setting and Modifying Alarm Sensitivity Levels Modifying Alarm Settings Configuring Email Notifications Cloning Agent Settings Setting and Modifying Alarm Sensitivity Levels Foglight for Oracle has three sensitivity levels that control which alarms are reported: • Essential — Enables only the most important alarms. • Normal — Enables a selection of alarms from different rule categories, including all Essential alarms. • Tuning — Enables all alarms. You can change the sensitivity level assigned to each agent. If a sensitivity level does not include all the alarms you want to track or includes too many alarms, you can view a list of multipleseverity rules and modify the sensitivity level that is mapped to each severity. Changes made to a sensitivity level affect all agents that are assigned that sensitivity level. If you want to enable or disable alarms for the selected agents, see “Enabling or disabling alarms for selected agents” on page 179. Managing Oracle Database Systems: User and Reference Guide 177 Chapter 2—Using Foglight for Oracle Setting the alarm sensitivity level by agent Each agent has its own sensitivity level setting. The default is Normal. Tip Select a sensitivity level that is closest to what you want, then customize it as necessary. For instructions, see “Viewing and modifying alarms assigned to severity levels” on page 177. To change the sensitivity level used by an agent: 1 In the Alarms view, click the Sensitivity Level tab. 2 Select the check boxes for agents you want to edit, and then click a sensitivity level button, such as Essential. 3 Click Save changes. Viewing and modifying alarms assigned to severity levels You can view a list of multiple severity rules to see which severities are assigned to which sensitivity level. If desired, you can change the assignments. Changes to sensitivity levels affect all agents. Tip If you want to see descriptions of the rules, follow the steps described in “Reviewing Rule Definitions” on page 187. To view and edit alarms assigned sensitivity levels: 1 In the Alarms view, click the Sensitivity Level tab. 2 Click View rules sensitivity mapping. Review the assignments. Recall that the Normal level includes all alarms assigned to the Essential level, and the Tuning level includes both Normal and Essential alarms. 3 If you want a record of the existing settings, click Report and export the settings to a file. 4 Select the check boxes for the rules you want to edit, and then click Set sensitivity. Managing Oracle Database Systems: User and Reference Guide 178 Chapter 2—Using Foglight for Oracle 5 In the Set Rule dialog box, select a sensitivity level for each severity. 6 Click Set. All selected rules are updated with the new assignments. Modifying Alarm Settings You can customize how the alarms generated by the default Foglight for Oracle rules are triggered and displayed in the Alarm view’s Settings tab. All changes to alarm settings apply to the selected agents, with the exception of thresholds, which can be customized by agent. Tip If you want to copy alarm settings to another agent that was not in the selected list— for example, you add a new agent — you can copy the alarm customizations to that agent. For instructions, see “Cloning Agent Settings” on page 186. Important Avoid editing Foglight for Oracle rules in the Administration > Rules & Notifications > Rule Management dashboard. Default rules may be modified during regular software updates and your edits will be lost. Always use the Alarms view. The Alarms list controls the contents displayed to the right and the tasks that are available. • All Alarms – Displays all rules with configured alarms and indicates whether alarms are enabled. In this view, you can enable or disable alarms for all the rules at once. You can also set email notifications and define mail server settings. • Category of rules – Displays a set of related rules with configured alarms. In this view, you can enable or disable alarms and also set email notifications for the category of rules. • Rule name – Displays the alarm status for the selected rule. If the rule has multiple severity levels, displays the threshold configured for each severity level. In this view, you can enable or disable the alarm, edit the alarm text, and edit severity levels and their thresholds. You can also set email notifications for the alarm. You can complete the following tasks: • Enabling or disabling alarms for selected agents • Modifying alarm threshold values • Editing the text of the alarm message Your changes are saved separately and applied over the default rules. This protects you from software upgrades that may change the underlying default rules. Managing Oracle Database Systems: User and Reference Guide 179 Chapter 2—Using Foglight for Oracle Enabling or disabling alarms for selected agents You can override the global alarm sensitivity level setting for the selected agents. You can enable or disable alarms for all rules, a category of rules, or an individual rule. Tip If you want to see descriptions of the rules, follow the steps described in “Reviewing Rule Definitions” on page 187. To enable or disable alarms: 1 In the Alarms view, click the Settings tab. 2 Decide on the scope for the change: all alarms, a category of rules, or a selected rule. 3 Complete the steps for the selected scope: Scope Procedure All alarms Click All Alarms. In the Alarms Settings tab, click either Enable all or Disable all. Category of rules Click a category. Click either Enable all or Disable all. Selected rule Click the rule. In the Alarms Settings tab, click the link that displays the alarm status. Select Enabled or Disabled from the list and click Set. 4 Click Save changes. Modifying alarm threshold values You can and should modify the thresholds associated with alarms to better suit your environment. If you find that alarms are firing for conditions that you consider to be acceptable, you can change the threshold values that trigger the alarm. You can also enable or disable severity levels to better suit your environment. When a rule has severity levels, a Threshold section appears in the Alarm Settings tab showing the severity levels and bounds by agent. For an example, see the DBO - OS General CPU Utilization rule. The threshold values corresponds to the lower bounds shown in this table. Many rules, such as Baseline rules, do not have severity levels and thresholds. When editing thresholds, ensure that the new values make sense in context with the other threshold values. For most metrics, threshold values are set so that Warning < Critical < Fatal. However, in metrics where normal performance has a higher value, such as DBO - SGA Library Cache, the threshold values are reversed: Warning > Critical > Fatal. Tip If you want to review the thresholds for all Foglight for Oracle rules in a single view, use the Rule Management dashboard. In the navigation panel, click Homes > Administration, then click Rules. Type DBO in the Search field to see the list of predefined rules for Oracle databases. For rules with severity levels, you can see the threshold values set for each level. If you want to edit Managing Oracle Database Systems: User and Reference Guide 180 Chapter 2—Using Foglight for Oracle threshold values, return to the Alarms view. Edits made directly to the default rules may be overwritten during software upgrades. To change severity levels and thresholds: 1 In the Alarms view, click the Settings tab. 2 Click the multiple-severity rule that you want to edit. 3 Click the Alarms Settings tab. 4 In the Threshold section, review the defined severity levels and existing threshold bounds for all target agents. The colored bar across the bottom of the view indicates the percentage of collections that fall within each threshold. 5 From this view, you can perform the following tasks: Task Procedure Edit severity levels and set threshold (lower bound) values for all agents. Click Enhance alarm. Select the check boxes for the severity levels you want enabled and set the threshold values. Click Set. Change the threshold (lower bound) values for one agent. Click Edit beside the agent name. Set the new threshold values and click Set. Copy the changes made to one agent’s threshold values to all other agents. Click Edit beside the agent name that has the values you want to copy. Select Set for all agents in table and click Set. 6 Click Save changes. Managing Oracle Database Systems: User and Reference Guide 181 Chapter 2—Using Foglight for Oracle Editing the text of the alarm message For individual rules, you can change the message displayed when an alarm fires. You cannot add or remove the variables used in the message. This is a global setting that affects all agents. To change the alarm message: 1 In the Alarms view, click the Settings tab. 2 Select a rule. 3 Click the Alarm Settings tab. 4 Click Enhance alarm. A Customize dialog box opens. 5 In the Message box, edit the message text. Tip If desired, you can restore the default message by clicking Reset message. 6 Click Set. 7 Click Save changes. Configuring Email Notifications We recommend that you set email notifications for the alarms you are most interested in tracking closely. For example, you may want to be notified by email of any Critical or Fatal situation. Or you may want to be informed whenever a key metric, such as CPU usage, is no longer operating within acceptable boundaries. You can set up email notifications that are generated when an alarm fires and/or on a defined schedule, as described in the following topics: • Configuring an email server • Enabling or disabling email notifications • Defining email notifications, recipients, and messages • Defining variables to contain email recipients • Defining scheduled email notifications Configuring an email server You need to define the global mail server variables (connection details) to be used for sending email notifications. To configure the Mail Server settings: 1 In the Alarms view, click the Settings tab. 2 Click All Alarms and then click the Email Server Settings tab. Note The Email Server Settings tab is available only when All Alarms is selected. 3 Click Edit. The Edit Mail Server Settings dialog box opens. 4 Type values for the following parameters: Managing Oracle Database Systems: User and Reference Guide 182 Chapter 2—Using Foglight for Oracle • • • • • • • Server Hostname — the name of the mail server Username — the username that is used for sending out email notifications Password — the password of the sender (that is, the username) Mail From — the sender’s email address Protocol — the mail protocol used for sending Email; either SMTP or SMTPS Port — mail server port Use SSL — defines whether to use secure connection 5 Click Set. 6 To ensure that the mail server is configured correctly, click Verify settings. The Verify Mail Server Settings dialog box opens. a Enter a valid email address. b Click Verify. If the verification fails, fix issues in the mail server configuration and repeat this process. 7 Click Save changes. Upon successful completion of the verification process, Foglight is configured for sending out email notifications. Enabling or disabling email notifications You can enable or disable email notifications for all alarms, a category of alarms, or a selected rule. Email notifications are sent only if all the following conditions are met: • The alarm email notification setting is enabled for the affected rule. • The alarm is triggered by changes in the monitored environment. • Alarm notification is enabled at the triggered severity level. See Defining email notifications, recipients, and messages. To enable or disable email notifications: 1 In the Alarms view, click the Settings tab. 2 Decide on the scope for the change: all alarms, a category of rules, or a selected rule. 3 Complete the steps for the selected scope: Scope Procedure All alarms Click All Alarms. In the Alarms Settings tab, click the Email notification settings button. Select either Enabled or Disabled from the Alarms notification status list. Click Set. Category of rules Click a category. Click the Email notification settings button. Select either Enabled or Disabled from the Alarms notification status list. Click Set. Managing Oracle Database Systems: User and Reference Guide 183 Chapter 2—Using Foglight for Oracle Scope Procedure Selected rule Click a rule. In the Alarms Settings tab, click the Email Notification Settings tab. Click the link that displays the alarm notification status. Select Enabled or Disabled and click Set. 4 Click Save changes. Defining email notifications, recipients, and messages You control who receives email messages, the subject line, and some text in the body of the email. The body of the email always contains information about the alarm. This information is not editable. You can also control whether an email is sent based on severity levels. You can set different distribution lists for different rules and different severity levels, or set the same notification policy for all rules. To configure email notifications: 1 In the Alarms view, click the Settings tab. 2 Decide on the scope for the change: all alarms, a category of rules, or a selected rule. 3 Complete the steps for the selected scope: Scope Procedure All alarms Click All Alarms. In the Alarms Settings tab, click the Email notification settings button. Continue to step 4. Category of rules Click a category. Click the Email notification settings button. Continue to step 4. Selected rule Click a rule. Click the Email Notification Settings tab. • To change the severity level that warrants an email notification, click the link that displays the severities. Select the desired level of severity and click Set. • To configure email recipients and the message, select the tab for a severity level, and click Edit. Skip to step 5. 4 If you selected All Alarms or a category, in the Email Notification Settings dialog box, do one of the following: • To change the severity levels that warrant an email notification, from the Messages will be enabled for severities box, select the desired levels of severity. • To configure the same email recipients and message for all severity levels, click Configure settings for all severities and then click All severities. • To configure different email recipients and messages for each of the severity levels, click Configure severity specific settings and then click a severity level. Managing Oracle Database Systems: User and Reference Guide 184 Chapter 2—Using Foglight for Oracle 5 In the Message Settings dialog box, configure the email recipients and message. • To — Type the addresses of the people who need to take action when this alarm triggers. • CC — Type the addresses of the people who want to be notified when the alarm triggers. Note If a mail server is not found, you are prompted to configure a mail server. For instructions, see “Configuring an email server” on page 181. Tip You can use registry variables in place of email addresses. Type the variable name between two hash (#) symbols, for example: #EmailTeamName#. For more information, see “Defining variables to contain email recipients” on page 185. • Subject — Optional. Edit the text of the subject line to better suit your environment. Avoid editing the variables, which are identified with the @ symbol. • Body Prefix — Optional. Add text that should appear above the alarm information in the body of the email. 6 Click Set to save the message configuration and close the dialog box. Managing Oracle Database Systems: User and Reference Guide 185 Chapter 2—Using Foglight for Oracle 7 If the Edit Notification Settings dialog box is open, click Set. 8 Click Save changes. Defining variables to contain email recipients You can create registry variables that contain one or more email addresses and (optionally) their scheduled notifications, and use these registry variables when defining email notifications. This procedure describes how to create a registry value. For schedules, see “Defining scheduled email notifications” on page 185. To create a registry variable: 1 On the navigation panel, under Dashboards, click Administration > Rules & Notifications > Manage Registry Variables. 2 Click Add. The New Registry Variable Wizard opens. 3 Select the registry variable type String, and click Next. 4 In the Name field, enter a name, for example: EmailTeamName Optional — Add a description. 5 Click Next. 6 Select Static Value. 7 In the Enter desired value box, enter one or more email addresses (separated by commas). Note Email groups are not permitted. 8 Click Finish. The Edit Registry Variable dashboard displays the newly created registry variable. To use a registry variable in email notifications, type the variable name between two hash (#) symbols, for example: #EmailTeamName#. For more information, see “Defining email notifications, recipients, and messages” on page 183. Defining scheduled email notifications If someone wants to receive an email about an alarm on a regular basis, such as once a day, you use a registry variable schedule to set up the notification. To schedule the sending of email notifications for a registry variable: 1 If you are continuing from Defining variables to contain email recipients, the registry variable is already open for editing in the Edit Registry Variable dashboard. Tip To edit a different variable, navigate to the Administration > Rules & Notifications > Manage Registry Variables dashboard, click the variable name, and select View and Edit Details. 2 In the Performance Calendars List table, click Add. The Performance Calendar Wizard opens. 3 Select a schedule, for example: End of Day 4 Click Next. Managing Oracle Database Systems: User and Reference Guide 186 Chapter 2—Using Foglight for Oracle 5 Select Static Value. 6 In the Enter desired value box, enter one or more email addresses (separated by commas) that should received email notifications based on the schedule. Tip The addresses may the same as or different from those assigned to the registry variable. 7 Click Finish. The Edit Registry Variable dashboard displays the newly created schedule. If desired, repeat to add other schedules. Cloning Agent Settings You may want an agent to have the same settings as another agent. For example, if you add new agents, you may want them to use the same settings as an existing agent. In this case, you can clone the settings from one agent to other agents. This process does not link the agents; in the future if you update the source agent, you also need to update the target agents. This procedure walks you through selecting the source agent from the Databases dashboard. However, you can also open the Administration dashboard with multiple agents selected. In this case, you select the source agent in Clone Alarm-related Settings to Other Agents dialog box. To clone alarm-related settings: 1 On the Databases dashboard, select the check box for the agent with the settings you want to clone. 2 Click Agent settings and then Administration. 3 In the Administration dashboard, click Alarms. 4 Click Set configuration on selected agents. The Clone Alarm-related Settings to Other Agents dialog box opens. 5 In the Select the source agent drop-down list, you should see the agent you selected. 6 In the Select the target agents table, select the check boxes for agents that should inherit settings from the source agent. 7 Click Apply. Managing Oracle Database Systems: User and Reference Guide 187 Chapter 2—Using Foglight for Oracle 8 When prompted for confirmation, click Yes. Reviewing Rule Definitions If you want to review the conditions of a rule, open the rule in the Rule Management dashboard. Important Avoid editing Foglight for Oracle rules in the Rule Management dashboard. These rules may be modified during regular software updates and your edits will be lost. Always use the Alarms view. You can create user-defined rules from the Rule Management dashboard. If you want to modify a rule, we recommend copying the rule and creating a user-defined rule. User-defined rules need to be managed from the Rule Management dashboard; these rules are not displayed in the Alarms view of the Databases Administration dashboard. For help creating rules, open the online help from the Rule Management dashboard. To open the Rule Management dashboard: 1 On the navigation panel, under Homes, click Administration. 2 In the Administration dashboard, click Rules. 3 Type DBO in the Search field to see the list of predefined rules for Oracle databases. The Foglight for Oracle rules are displayed. From here, you can review threshold values, alarm counts, and descriptions. 4 To see the full rule definition, click a rule and then click View and Edit. 5 In the Rule Detail dialog box, click Rule Editor. 6 When you are done your review, click Rule Management in the breadcrumbs to return to the dialog box. Defining Retention Policies Use the Retention Policies view in the Databases Administration dashboard to define the requested duration for which each of the metric collections, which are defined in the Collections view, are to be kept. Data can be retained for brief, moderate, or long periods, by selecting one of the following options: • Short — retains data up to one month. • Medium — retains data up to three months. • Long — retains data up to one year. To modify the retention policy: 1 Click the text that displays the current retention policy. The Edit the Retention Policy dialog box opens. 2 Select the requested retention policy scheme from the list. 3 Click Set to apply the selected setting or Cancel to reject the setting. 4 At the bottom of the view, click Save changes. Managing Oracle Database Systems: User and Reference Guide 188 Chapter 2—Using Foglight for Oracle Defining the Collection Settings Use the Collection Settings view to limit the number of displayed items for each of the collections below. Locks Use the Locks pane to define the following settings: • The maximum number of blocked segments to be displayed in the Lock Overview pane of the Activity > Locks panel. • The minimal duration, in seconds, which a lock should reach or exceed in order to be collected. All locks that meet this criterion are displayed in the lower section of the Locks Historical Tree pane. To modify the lock collection and display settings: 1 Click Edit. The Edit Lock Collection and Display Settings dialog box opens. 2 Use the designated field to type the maximum number of blocked segments. 3 Use the designated field to type the minimal duration for collecting and displaying a lock in the Locks Historical Tree 4 Click Set to approve the new setting or Cancel to undo the operation. Top Sessions Use the Top Sessions pane to set limits to the maximum retrieved number of top sessions, that is, sessions that experienced the highest contention as per the selected criterion. All of the retrieved sessions are displayed on the Top Sessions tab of the Sessions List section of the Activity > Sessions panel. To define the maximum number of top sessions to be retrieved: 1 Click Edit. The Edit Top Sessions Settings dialog box opens. 2 Enter the maximum number of sessions in the designated field. Alternatively, choose to retrieve all sessions. 3 Use the Contention Criterion drop-down list to select the field by which the Top Sessions list is to be sorted. The possible values are: physical reads, physical writes, sort disk, logical reads, logical writes, logon time, database time, and database CPU time. 4 Click Set to approve the new setting or Cancel to undo the operation. Top SQL Statements Use the Top SQL Statements pane to configure the requested settings for top SQL statements, that is, statements that generated the maximal workload, as defined by the criterion by which the selected SQL statements are sorted. Use this pane to define the criterion of top SQL statements, as well as the maximum number of such statements to be retrieved and displayed. Top SQL statements are displayed on the Activity > Top SQL Statements panel. Managing Oracle Database Systems: User and Reference Guide 189 Chapter 2—Using Foglight for Oracle To define the settings for collecting and displaying SQL statements: 1 Click Edit. The Edit Top SQL Statements Settings dialog box opens. 2 Use the Maximum number of displayed SQL statements field to enter the maximum number of SQL statements that are to be displayed in the Top SQL Statements grid. 3 Use the Maximum number of retrieved SQL statements field to enter the number of SQL statements that generated the maximal workload and are to be saved to the Top SQL statements collection. 4 Use the Sort the collected SQL statements by list to select the field by which the Top SQL Statements list is to be sorted. The possible values are: • CPU Time — the total CPU time consumed for carrying out the SQL statement executions • Elapsed Time — the total time consumed for carrying out the SQL statement executions • Executions — the number of times the SQL script executed for a unique SQL Handle • Total Waits — the total wait time spent while executing the SQL statements 5 Use the Maximum size of short SQL statement field to type the maximum number of characters for the short SQL text. 6 Click Set to save these settings. 7 At the bottom of the view, click Save changes. Tablespace Settings Use the Tablespace Settings pane to configure the default monitoring settings for the Tablespaces Storage collection, by excluding from monitoring one or more of the following types of tablespaces: • Undo tablespaces • Temporary tablespaces • Read-only tablespaces • System-Sysaux tablespaces This pane also includes the Tablespace Exclude List section. Use this section to add one ore more tablespaces to the Exclude list. To add tablespaces to the Exclude list: 1 Click Add. The Add a Tablespace to the Exclude List dialog box opens. 2 Use the Tablespace name field to enter a field name. To exclude multiple tablespaces, insert a regular expression (wild card) and select the Regex check box. 3 Click Add to complete the operation. Managing Oracle Database Systems: User and Reference Guide 190 Chapter 2—Using Foglight for Oracle Invalid Objects Settings Use the Exclude List pane to excluding specific schemas from being monitored by the Invalid Objects collection. To exclude a schema from being monitored: 1 Select the agent that monitors the instance where the schema resides. Important If the selected agent is a RAC node, changes made will apply to RAC agent, thereby affecting the entire RAC. 2 Click Add a Schema to the Exclude List. 3 Use the dialog box that appears to enter the name of the requested schema. 4 Click Add. 5 Repeat this procedure for all of the schemas that are to be excluded. To remove schemas from the table, select the schema’s row and click Delete. 6 At the bottom of the view, click Save changes. Usability Response Time Use the Usability Response Time pane to edit the query used for measuring the instance's average response time, that is, the full time (in milliseconds) it has taken this query to get from the application to Oracle and back. Note The response time cannot be edited for RAC nodes, but only for RAC and single instance agents. If all of the agents are selected before entering the Databases Administration dashboard are RAC nodes, no change will take place. If some of the selected agents are RAC nodes, the response time will be partially modified. Configuring the Alert Log Panel Display Use the Alert Log view to configure the retrieval and display settings for the Alerts > Alert Log panel. In addition, this view is used to define default settings for invoking alarms based on the collected alerts. Important While alerts whose severity type is Informational are collected and displayed on the Alert Log panel, alarms are invoked only for alerts whose severity type is Warning or higher. Use this view to define the following settings: • Minimal severity for invoking summary alarms — can be either turned off or set to one of the defined severity values, that is: Warning, Critical or Fatal. Note By default, the value of this parameter is Critical. • Minimal severity for invoking alarms — can be either turned off (the default setting) or set to one of the defined severity values: Warning, Critical or Fatal. Managing Oracle Database Systems: User and Reference Guide 191 Chapter 2—Using Foglight for Oracle • Number of rows to collect from logs — the maximum number of rows that would be retrieved from the alert logs. • Display under a pre-defined name in the Alert Log panel — using the Match List pane. This pane contains a default list of expressions within the alert logs that are to be retrieved and displayed in the Alert Log panel under a pre-defined name, category and severity. • Exclusion from the Alert Log panel display — using the Ignore List pane, which contains a default list of alerts that are to be excluded when setting the alert logs display. Important Only messages that are explicitly defined in the Ignore List will not be displayed. Messages that were not added to either the Match or Ignore lists appear under name Other, type Oracle alert messages and severity Informational. Therefore, ensure that messages that need not be displayed are added to the Ignore List. Both the Match List and the Ignore List panes can be customized by adding, editing or removing alert logs. Each filter can be enabled or disabled separately by clicking Edit and selecting or clearing the Enabled check box. Alternatively, to enable or disable all of the filters, click the Enable All or Disable All button. To add an alert to the Match List: 1 Click Add. The Add an Alert to the Match List dialog box opens. Use this dialog box to configure the alert filter by means of the following fields: • Filter Name — used for assigning a name to the filter. • Severity — used for assigning a severity for the alert. Use this field to set the type of the alarm that is invoked as a result of the alert, or to define that the alert is informational and therefore invokes no alerts. • Type — used for entering a filter type, either one of the predefined types or a new type. • Expression — identifies the text that is to be extracted as a message from the alert log and displayed in the Message column of the Alert Log panel > Alert Log Messages table. Managing Oracle Database Systems: User and Reference Guide 192 Chapter 2—Using Foglight for Oracle This text usually contains the beginning of the message and a wildcard (for example: ORA-12012.*). Note Alert filters are enabled automatically upon addition. To disable a filter, use the Edit button. 2 Click Add to save your settings. Each newly added alert filter is enabled by default. To disable the filter, click Edit and then clear the Enabled check box. To edit an alert: 1 Click the requested alert log. 2 Click Edit. The Edit the Selected Alert dialog box opens. 3 Use this dialog box to configure the alert by means of the same fields used for adding it: Filter Name, Severity, Type, and Expression. 4 To disable the filter, clear the Enabled check box. 5 Click Set to save your settings. 6 At the bottom of the view, click Save changes. Configuring the List of Listeners per Instance Use the Listeners view in the Databases Administration dashboard to edit the list of monitored listeners per instance, by adding or removing listeners from the list. To add listeners to an instance: 1 Open the drop-down list below the title Please select an agent:, to display the entire list of selected agents. Note As RAC agents whose version is earlier than 11G2 have no listeners, they are not included in the list, even if such agents appear in the list of selected agents. 2 Select a single agent from the list. 3 Click Add Listener. The contents of the Add a Listener dialog box that appears now vary between Windows and UNIX. For instructions for Windows-based agent, go to step 4. In UNIX, enter a listener service name in the designated field. 4 For a Windows-based agent, type data in the following fields: • Listener name — the name of the listener • Oracle home — the path used for installing the listener • TNS Admin — the path of the $TNS_ADMIN directory (usually $ORACLE_HOME/ network/admin) 5 Click Add. The newly added listener appears in the Listeners table. Managing Oracle Database Systems: User and Reference Guide 193 Chapter 2—Using Foglight for Oracle To edit the properties of a single listener: 1 Select the requested listener from the table. 2 Click Edit Listener. 3 In the Edit a Listener dialog box, edit the existing data in the various fields (see step 4 of the procedure for adding a listener). 4 Click Set to apply the changes or Cancel to revert to the previous settings. To stop monitoring a listener: 1 Select the requested listener from the table. 2 Click Delete Listener. The Stop Monitoring a Listener dialog box opens. 3 Click Yes to confirm the deletion or Cancel to keep monitoring the listener. 4 At the bottom of the view, click Save changes. Configuring the Connection to Performance Analysis Use the Performance Analysis view to maintain the connection to Performance Analysis, a Quest technology designed to collect SQL statements and performance indicators from Oracle databases and host operating systems using memory sampling, without imposing overhead on the target database itself. This view displays only the agents whose connection to Performance Analysis was configured using one of the following methods: • Editing the agent credentials in the Connection Details screen. Edit these credentials by enabling the connection to Performance Analysis and configuring the connection details, thereby setting user-defined connection credentials. • Using the Detect Performance Analysis button in the Connection Details screen in the Databases Administration dashboard. When using this method, the connection is carried out using the default connection details, except for the optional modification of the Performance Analysis port. This view contains the Performance Analysis Administration table, which displays the following columns for each monitored Performance Analysis agent: • Agent name • The status of the agent’s retrieval components — Collector, Middleware and Repository database; either Up (running) or Down • Details about the agent Use the Performance Analysis view in the Databases Administration dashboard to monitor the agent functioning and, if needed, carry out one or more of the following maintenance operations: • Stopping and restarting the agent by clicking Deactivate Performance Analysis and Activate Performance Analysis. Managing Oracle Database Systems: User and Reference Guide 194 Chapter 2—Using Foglight for Oracle • Validating the agent’s connectivity using the Verify accounts button. • Generating the log information for contacting Quest support, by clicking Generate support bundle. • Using the Set connection profile button to modify the agent’s credentials and the Performance Analysis repository connection details. Configuring User-defined Collections Use the User-defined Collections view in the Databases Administration dashboard to add userdefined collections to all of the currently selected agents, in order to provide for queries not included in Foglight for Oracle. After collections are added, this view displays all user-defined collections for all of the agents; for example, if a collection was added to 12 agents during its addition, the view will display 12 rows, showing the collection for each agent. This view is also used for configuring the sampling frequency for each collection. The available sampling frequencies are: • Real-Time - When a user is currently focusing on a screen, the sampling frequency for all of the collections associated with this screen switches to Real-Time. • Online - When at least one user is connected, the sampling frequency for all of the collections that are not currently running at Real-Time frequency switches to Online. • Offline - when no user is currently connected to the application To add user-defined collections: 1 Select the requested agents. 2 Click Add. If the user-defined collections are enabled for the selected agent, the Add a User-defined Collection dialog box opens. Proceed to step 3. If user-defined collections are not enabled, the User-defined Collections Not Enabled dialog box opens. Important Agents can also enabled for user-defined collections through the Connection Details view. For details, see “Enabling and setting credentials for user-defined collections” on page 170. To enable and configure the creation of user-defined collections: a Click Edit credentials. The Edit Credentials for User-defined Collections dialog box opens. b Select whether to perform the collection using the existing agent credentials or by creating dedicated credentials. c If the user-defined collections are to be enabled using dedicated credentials, type the requested user name and password. d Select whether to connect as SYSDBA. Managing Oracle Database Systems: User and Reference Guide 195 Chapter 2—Using Foglight for Oracle e Click OK. f The Applying modified settings progress bar is displayed. If the modified settings are su cess fully applied, the Add a User-defined Collection dialog box opens. 3 Click Set details. The Set Collection Details dialog box opens. 4 Enter a name in the Collection name field. 5 Enter a brief description of the collection in the Collection description field (optional). 6 Paste the query’s SQL text in the Query Text field. Note This field can hold up to 4096 characters. Any PL-SQL Functions (that are not getting input) can be written by wrapping the PL-SQL "Select From dual" statement. 7 Type a value, in seconds, in the Query Timeout field. 8 Use the Start Time field to define the time when the collection starts running (optional). Note If this field is left empty, the collection starts working immediately. The start time can only be defined using the wizard. After the collection starts working, at the time defined using the Start Time field, it will continue to run according to its defined frequency and can only be disabled. 9 Click Verify. The collection is verified by running the query on each of the currently selected agents. After the verification process is complete, the Verification Results pop-up appears, indicating whether the collection was verified successfully. In case the collection verification failed, the relevant error message is displayed. If the verification succeeded on at least one agent, the collection’s details are displayed on the table described below. Managing Oracle Database Systems: User and Reference Guide 196 Chapter 2—Using Foglight for Oracle Section Field Description Database ID The collections’ ID Column Name The name of the column Column Type The field type, as retrieved by the query (String, Integer and so on) Display Name The column’s display name Type The topology type for storage purposes. This type can be one of the following: • String • Integer • Double • Boolean • Date Data Storage Note When the Frequently modified check box is selected (the default state), the field’s change history is kept, including use of optional functionalities such as use of the IntelliProfile mechanism and aggregation type selection. Clearing this check box is recommended only for fields whose values change infrequently, such as IP address of a specific host, as storing the change history of such fields is highly CPU-intensive and may degrade the FMS performance. Unit of Measurement The metric’s unit of measurement. The possible measurement unit values are: Percent, Count, Millisecond, Second, Minute, Hour, Day, Byte, Kilobyte, Megabyte and Gigabyte. Aggregation Used for defining the value that is displayed in this metric, out of several values that were returned in the relevant time range. The available aggregation values are: • As is — value of the last sample taken during the selected time range • Sum — summarized value of all samples taken during the selected time range • Average — average value of all samples taken during the selected time range Managing Oracle Database Systems: User and Reference Guide 197 Chapter 2—Using Foglight for Oracle Section Field Description Is Key Indicates whether the field is the query’s key for retrieval. Note Configured to use IntelliProfile Fields that are indicated as keys should correspond with the database result set unique values. Selecting a field which has frequently changing and repeating results as key is allowed, but may result in unexpected behavior from the defined collection. Select True in this column to enable the collection of baseline statistics for this metric. Except for the Column Name and Column Type fields, whose values are retrieved by the query and cannot be changed, all other fields can be edited by clicking any of them. The Edit Collection Properties dialog box appears, allowing you to edit the values of the following parameters: • • • • Display name Type Unit of measurement Aggregation In addition, the collection’s sampling frequencies are displayed on the table at the bottom of the screen, and can be edited by clicking any of them. 10 Ensure that all settings are appropriate, and click OK to finish the collection creation process. The newly created collection now appears on the table. Important After adding the requested user-defined collections, they can only be deleted or cloned to other agents. If one or more queries need to be modified, delete them and create new ones. Configuring the On-demand Data Port Use the On-demand Data Port view in the Databases Administration dashboard to define a port for each Foglight Agent Manager to be used by for retrieving data and integrating with Performance Analysis. To configure a port for the collection of on-demand data: 1 Select the check box near the names of the requested Foglight Agent Managers. 2 Click Set port. The Set On-demand Data Port dialog box opens. 3 Type the name of the requested port. Managing Oracle Database Systems: User and Reference Guide 198 Chapter 2—Using Foglight for Oracle 4 Click Set. 5 Click Validate connectivity. The Validate connectivity progress bar appears. At the end of the validation process, the connection status appears on the Status column. If the connection failed, take the requested correction measures and try again. Important Changes to the On-demand Data Port view take effect immediately and do not need to be saved. Changing the Monitoring Mode of the RAC One Node Agents Use the RAC One Node view in the Databases Administration dashboard to change the monitoring mode of agents configured as RAC One Node, enabling the monitoring of such agents as regular RACs. Note The RAC One Node view is only relevant for agents configured as RAC One Node. If no such agent was selected before entering the Databases Administration dashboard, this view will be empty. To change the monitoring mode of a RAC One Node agent: 1 Click anywhere on the row of the RAC One Node agent (the root). The Convert to Regular RAC dialog box appears, displaying one or more hosts. 2 Click the row of the RAC that is to be converted to regular RAC. The Convert to Regular RAC dialog box opens. 3 From the list of hosts that appear in the dialog box, select the host that is to be converted to a physical RAC. 4 Click Validate to verify the selected host’s connectivity. 5 After the connectivity validation completes successfully, click Set. Note Converting a RAC One Node to a regular RAC is an irreversible process. The Save changes button is disabled on the RAC One Node view, and after clicking Set, the selected host is immediately converted to a physical RAC. Therefore, ensure that the requested host is selected before clicking Set. After the selected host was converted to a physical RAC, use the Oracle Monitoring Installer wizard to add the other hosts monitored through this agent as nodes to the newly created RAC. For further details, see “Running the Oracle Monitoring Wizard” on page 24. Managing Oracle Database Systems: User and Reference Guide 199 Chapter 2—Using Foglight for Oracle Generating Reports Foglight for Oracle enables generating reports about various aspects of the selected instance or RAC performance. This chapter provides information on how to generate the various reports, as well as a brief description of each report. Note For detailed information regarding the use and configuration of reports, see Foglight User Guide > Working with Reports. Generating Reports for a Foglight for Oracle RAC or Instance To generate reports for a selected RAC or instance: 1 Go to Dashboards > Reports. 2 Click Run a Report. The Run Report dialog box opens. 3 On the first screen, Select Template, Go to All Oracle Templates. 4 Select the requested report. 5 Click Next. 6 Select the requested time range. 7 Select the instance or RAC for which the report will be generated. 8 Click Next. 9 Assign a name for the report. 10 Select the report format (PDF, Excel or XML). 11 Use the Email Recipients field to type the name of the report’s recipients. 12 Click Finish. The report generation process starts. Upon successful completion of this process, the Report Generated Confirmation dialog box opens. 13 Click Download Now to download the file. 14 Open the file using a relevant program for the selected format. Studying the Various Reports The reports that can be generated for Foglight for Oracle are described below. Except for the Cluster Balance report, all of the reports can be generated for either a RAC or an instance. • Cluster Balance report — used for monitoring the load balance, during the selected time range, between the various nodes in the selected RAC. • Databases Storage report — displays a breakdown of the database storage space usage at the host, instance and tablespace levels. If the monitored instance is part of a multitenant Managing Oracle Database Systems: User and Reference Guide 200 Chapter 2—Using Foglight for Oracle architecture, this report at the instance level includes a graph of pluggable database distribution by storage. • Health Check report — used for monitoring various aspects of the instance’s or RAC’s health, that is: availability, listener status, response time and connection time. • Host Disk Capacity Report — provides information about the file system disk of each of the monitored host, such as the file system type, the disk size, the amount (MB) of used and free space on the disk, and the percentage of free space. • I/O Activity report — provides several I/O activity indicators, such as workload, physical and logical read and write operations, and wait events. • Instance Availability Report — provides storage-related information about the instance, RAC node or cluster, such as availability percentage, listener status, connection time, and average response time. • Oracle Executive Summary report — provides key decision-makers with a summary view of the Oracle activity. If the monitored instance is part of a multitenant architecture, this report displays a grid showing data for each pluggable database. • PL/SQL Blocks report — displays the PL/SQL blocks that generated the highest amount of activity for a given time frame. • Storage Summary report — provides various storage-related indicators, such as the status of datafiles and ASM, and the top 10 tablespaces sorted by space usage. If the monitored instance is part of a multitenant architecture, this report at the instance level includes a graph of pluggable database distribution by storage. • Top DB Users report — displays the DB users that generated the highest amount of activity for a given time frame. • Top SQLs report — provides detailed information about the SQL statements that experienced the longest time of CPU consumption or highest number of total wait events during the specified time range. The details of these statements are displayed both in a bar chart and in a table. • Workload Summary Report — provides various workload indicators, such as the workload resource breakdown, CPU utilization and SQL executions. Glossary Alarm The mechanism by which Foglight for Oracle alerts users to a condition that might be a problem in the Oracle instance. Authentication The process of identifying and verifying a user who is attempting to establish an Oracle session. With Oracle Authentication, users must have an Oracle login and password which is validated against an Oracle system table. Managing Oracle Database Systems: User and Reference Guide 201 Chapter 2—Using Foglight for Oracle Calibration The process by which Foglight for Oracle determines the maximum and minimum values for every dataflow on the home page, by observing data moving through the database system. This information helps Foglight for Oracle display the dataflows correctly. Cartridge Cartridges extend the functionality of Foglight and are installed on the Foglight Management Server. A cartridge contains one or more components, such as agents for deployment, communication capabilities, and modifications to the way that data is transformed or handled, as well as rules, reports, and views. When a cartridge is installed and enabled, its components become part of the Management Server. Adding cartridges allows users to monitor additional parts of their environment. For further details about managing cartridges, See the Foglight Administration and Configuration Guide. Cursors Extensions to result sets that provide the mechanism for working with individual rows, or a small block of rows, in a table. Because a cursor points to a currently selected set of records, they can be used by only one connection at a time. However, the compiled plan to which the cursor is linked can be used simultaneously by multiple connections. DataGuard Oracle Data Guard ensures high availability, data protection, and disaster recovery for enterprise data. Using Data Guard’s set of services for creating, maintaining, managing, and monitoring one or more standby databases, enables production Oracle databases to survive disasters and data corruptions. These standby databases are then maintained as transactionally-consistent copies of the production database. If the production database is unavailable due to outage (either planned or unplanned), these copies enable Data Guard to switch any standby database to the production role, thereby minimizing the downtime associated with the outage. DataFlow A Dataflow displays the current level of activity. As the rate of data transfer increases, so too does the speed of the flow. If the statistic that the flow represents moves to another threshold, the flow may change color. The combination of movement and color makes it easy to spot congested areas. A graph above the flow shows how the load has varied over time. Deadlock A deadlock occurs when two or more session are waiting for data locked by each other, resulting in a situation where all sessions are being blocked. Oracle automatically detects and resolves deadlocks by rolling back the statement associated with the transaction that detects the deadlock. Managing Oracle Database Systems: User and Reference Guide 202 Chapter 2—Using Foglight for Oracle Disk Transfer Time Reading or writing data requires a disk to access the disk sector where the requested data resides. After this sector is accessed, the amount of time required for a disk to read or write data from or to storage media is referred to as disk transfer time. Transfer time, usually expressed in milliseconds, is part of the disk access time, that is, the total time required for the computer to process the data request from the processor and then retrieve the needed data from a storage device. Disk Utilization The percentage of elapsed time during which a disk is busy servicing I/O requests. DiskPerf A Windows command line utility that enables or disables the collection of I/O statistics. Dispatcher Dispatcher is a background process, which is responsible for routing requests from connected user processes to available shared server processes, and for returning the responses back to the appropriate user processes. The dispatcher process, which is only utilized with shared server configuration, handles and directs multiple incoming network session requests to shared server processes. At least one dispatcher process has to be created for every communication protocol available on the server. Drilldown A Foglight for Oracle view that provides more detailed information about a particular element (for example, a monitored object or an alarm) than the current view or dashboard. Foglight for Oracle drilldowns often contains charts or tables showing Oracle or Windows statistics or objects. External Procedures An external procedure, also sometimes referred to as an external routine, is a procedure stored in a dynamic link library (DLL) on Windows or shared library under UNIX. The external procedure is registered with the base language and then invoked to perform special-purpose processing. Because external procedures run in a process separated from the database instance, using these procedures ensures that any problems on the client side do not adversely affect the database. Foglight Agent Manager The Foglight Agent Manager is a client application that manages Foglight agents installed on monitored hosts. It provides a centralized communications link between the Foglight Management Server and the agents. The Foglight Agent Manager also provides a number of support services such as the ability to deploy, upgrade, and configure agents. For further details, see the Foglight Agent Manager Installation Guide. Managing Oracle Database Systems: User and Reference Guide 203 Chapter 2—Using Foglight for Oracle Foglight Management Server The Foglight Management Server (FMS) is the central component of Foglight. The Management Server receives information from agents and makes it available in the browser interface. The Foglight database stores all system, application, and performance data. Over time, it becomes an invaluable source of historical information for planning future system capacity requirements and for doing point-in-time analysis. Host Name The name of the client computer that established the Oracle connection. This name appears in the Foglight for Oracle Sessions, Locks, and Blocking views. Host Process A string that uniquely identifies the instance of the client application that established the Oracle connection. This string appears in the Foglight for Oracle Sessions, Locks, and Blocking views. Kernel Memory The physical memory allocated to Windows kernel. Kernel Mode See “Privileged Mode” on page 205. Kill A Transact-SQL statement that terminates an Oracle connection. Any outstanding transactions for the selected session are rolled back and all locks are released. LRU Least Recently Used. An algorithm used by Oracle for making room for new information in the memory space allocated, by replacing the oldest (least recently used) data. LSN Log Sequence Number. This number, which uniquely identifies a point in a database's Log, is used for determining when a page was last modified. Licensing The legal agreement that determines how many users can access each Oracle installation. See glossary definitions of “Per-Seat Licensing” on page 204 and “Per Server Licensing” on page 204. Logical Reads The number of data blocks read by Oracle, which were already present in the buffer cache, thereby saving input/output read operations. This value is also referred to as session logical reads and is the total of Current Reads plus Consistent Reads. Managing Oracle Database Systems: User and Reference Guide 204 Chapter 2—Using Foglight for Oracle Metric A unit of measurement that can be applied to a database. Metrics can help gauge the performance of a system. A metric is an individual piece of information that Foglight for Oracle collects about the performance of a system. The information may be a numeric value (a number or percentage), a string of text, or some other piece of data. Every time the Foglight for Oracle dashboard is refreshed, the cartridge retrieves the latest value of the metric, which can then be displayed in a drilldown or on the home page. OLTP Online Transaction Processing. OLTP allows real-time processing of SQL transactions, in order to support Customer Relationship Management (CRM), ERP, and other time-critical applications. OLTP is characterized by high rates of index lookups, single row modifications, and frequent commits. Because real-time transaction processing is being increasingly carried out on a network and may include more than one company, OLTP databases use client/server processing and allow transactions to run on different platforms in a network. Paging Disk I/O activity performed by the operating system to manage its virtual memory. High paging rates can adversely affect performance. Panel A group of related components on the Foglight for Oracle screens. Per-Seat Licensing The Oracle licensing mode that allows multiple instances of Oracle to be deployed, and allows specific users to access all of them. See “Licensing” on page 203 and “Per Server Licensing” on page 204. Per Server Licensing The Oracle licensing mode that allows a stated maximum number of simultaneous users to be supported for a specific installation of Oracle. See “Licensing” on page 203, and “Per-Seat Licensing” on page 204. Physical Reads The number of data blocks read by Oracle, which are not present in the buffer cache or in the PGA (Program Global Area) memory, resulting in an I/O read operations. While physical read operations are inevitable in certain situations, they should generally be avoided. Managing Oracle Database Systems: User and Reference Guide 205 Chapter 2—Using Foglight for Oracle PID Process ID. A unique number that identifies a Windows process at any given point in time. Privileged Mode An operational state of hardware or software that has the highest priority. This mode is primarily used by operating system services such as I/O and paging. Also known as Supervisor Mode and Kernel Mode. See “User Mode” on page 206. Process An instance of an application executing in Windows. Random I/O I/O in which a specific disk block is directly accessed. This is typically the I/O that results from index lookups. Recompile The process of compiling a stored procedure part way through that procedure's execution. Referential Integrity Referential Integrity ensures that foreign keys correctly map to primary keys. A referential constraint prevents the insertion or update of foreign keys for which there are no matching primary keys. It either prevents the deletion of primary keys if foreign keys exist, or deletes these foreign rows (DELETE CASCADE). Relational Data Engine A major functional part of Oracle. Responsible for the parsing and optimization of SQL requests, controls query plan execution, and processes row sets from the storage engine. Scattered Read I/O operation in which multiple disk blocks are read and scattered into different discontinuous buffers in the SGA. This is typical of the I/O that results from performing operations that require processing large amounts of data, such as full table scans, sorting, or Oracle Parallel Queries (OPQs). Schema Locks A lock mode that is used when compiling a query, to prevent changes to the underlying tables structures while the query is compiled. Sequential Read I/O operation in which disk blocks (usually single blocks) are read in sequence into contiguous memory space. This is typical of the I/O that results, for example, from using indexes. Managing Oracle Database Systems: User and Reference Guide 206 Chapter 2—Using Foglight for Oracle Severity Represents how critical an alarm is. A severity determines how Foglight for Oracle behaves when the values for a metric fall within a user-defined range of values. A severity specifies whether the information returned in the metric represents normal or abnormal behavior for the system under diagnosis. For example, unusually high values might mean that a metric has crossed a threshold into a high severity state. This, in turn, could change the color of a component on the home page; for example, from orange to red when moving from a critical to a fatal state. The severity determines what action Foglight for Oracle takes when a metric value falls into the range defined by a threshold. Spinner A Spinner displays the current level of activity for a statistic that is not directional. As the load increases, so too does the speed of the spin. If the statistic represented by the flow crosses another threshold, the spinner may change color. The combination of movement and color makes it easy to spot congested areas. SQL *net SQL *net is the Oracle network interface that allows Oracle tools running on network workstations and servers to access, modify, share, and store data on other servers. SQL *net is considered part of the program interface in network communications. Threshold A range of values for a metric. If the metric falls within this range, Foglight for Oracle checks the threshold's severity to determine how to behave. For example, the component that represents this metric might change color. Transaction A group of one or more database modification statements that are combined into a logical unit of work that is either wholly committed or rolled back. Unused Space Disk space within a Database or File Group that is allocated to a table or index, but currently does not have any information stored in it. It is free space that can only be used by the table or index to which it is allocated. User Mode A processor mode that is primarily used for executing application code. Unlike the Privileged (Kernel) Mode, processes running in the user mode do not interact directly with the operating system’s kernel, memory, or hardware directly. As a result, whereas a crash of a kernel mode process in most cases leads to the entire system’s crash, a crash of a user mode process usually only terminates the application that was run using this mode. See “Privileged Mode” on page 205. Managing Oracle Database Systems: User and Reference Guide 207 Chapter 2—Using Foglight for Oracle Performance Analysis Metrics Active Time Sum of all active wait events; equal to the session total activity within the specified time range. Agent Down (%) The amount of time that the Performance Analysis agent was not running (down) during the specified time range. Archive Wait Any activity that generates Log WRiter (LGWR) for archiving the redo logs to the target location. Additionally, user sessions can also experience archival waits when the log that the LGWR will be switching into has not yet been archived and the session is issuing DML statements. Average Collisions Per Packet Average number of collisions per packet during the specified time range. Average Lock Duration The average time (in milliseconds) during which a resource was locked (held in contention). High values of lock duration may indicate aggressive lock incidence, which may be improved through tuning. The metric is calculated as follows: [100 * (Lock Wait)]/Enqueue Waits Average Physical I/O Operations The average time spent on performing physical I/O operations (both read and write), out of the total active time during the specified time range. This metric is calculated as follows: (Page Reads + Page Writes)/Active Time Average Read Time The average time (in milliseconds) spent reading a data block from the disk into the database buffer cache. Average Redo Write Time The average time (in milliseconds) spent writing a redo log entry to the log files. User sessions experience log write wait events when, upon commits, an individual session waits for the LGWR to flush its redo information to the redo log file. Average Write Time The average time (in milliseconds) spent writing a data block from the database buffer to the disk. Managing Oracle Database Systems: User and Reference Guide 208 Chapter 2—Using Foglight for Oracle Average SQL Response Time Average duration of the SQL statements that were executed during the specified time range. This metric is calculated as follows: Ended Duration/Executions Ended Backup Recovery Wait Time spent by the various sessions waiting for backup/recovery tasks to complete. When archiving is not performed fast enough, database operations can be suspended temporarily whenever the log writer is forced to wait for an inactive group to become available for reuse. Block Changes The total number of changes to blocks that resulted from update or delete operations. Block Reads This metric determines the block get rate, which is a basic measure of the rate at which the application system references the database. Change in this rate can indicate either that the database workload has significantly increased, or that the referenced segment sizes increased. Buffer Busy Wait The time spent by the session when trying to access a data block that is unavailable for one of the following reasons: • The data block is currently not in memory, but another process has already issued an I/O request to read it into memory • The data block is in memory, but in an incompatible mode (for example, current instead of consistent) Buffer Wait The time spent waiting for the completion of buffer operations. This wait event usually results from insufficient or non-optimal buffer configuration. Calls The summarized value of user calls and recursive calls. For details, see “Recursive Calls” on page 218 and “User Calls” on page 221. Cache Hit Ratio This metric measures the effectiveness of the buffer cache. The metric is calculated as the percentage of block requests that found the block in the buffer cache, without requiring disk access, out of the entire block requests (including requests that required disk reads). Chained Row Ratio This metric measures the number of chained or migrated rows encountered during a fetch, as a percentage of the entire fetched rows. Managing Oracle Database Systems: User and Reference Guide 209 Chapter 2—Using Foglight for Oracle Consistent Changes The number of times a user process applied rollback entries to perform a consistent read on the block. Consistent Gets The number of data block reads in Consistent mode (that is, consistent reads). A consistent read retrieves the block version at the time the query has started. Consistent mode reads are usually associated with query activity. Control File IO Wait Time spent waiting for the completion of input/output requests, reads, and writes to control file blocks. CPU Usage When SQL statements and other types of calls are made to Oracle, an amount of CPU time is necessary to process the call. The average call requires a small amount of CPU time. However, a SQL statement involving a large amount of data or a runaway query can potentially consume a large amount of CPU time, reducing CPU time available for other processing. CPU utilization is used as a key operating system statistic in the tuning process. Excessive CPU could be caused by an inadequately-sized system, by untuned SQL statements, or by inefficient application programs. CPU Wait Time spent by the session waiting in the system's run queue to be granted for CPU cycles. The amount of time is dependent upon the number of concurrent processes and threads requesting CPU time. Cursor Synchronization Wait The time spent by the various processes synchronizing information flow within cursors. Cursors are extensions to result sets, which provide the mechanism for working with individual rows, or a small block of rows, in a table. As a cursor points to a currently selected set of records, a cursor can be used by only one connection at a time. However, the compiled plan to which the cursor is linked can be used simultaneously by multiple connections. DB Block Gets The number of data block reads in Current mode (that is, current reads). A current read retrieves the latest version of the block at the time of reading. Current mode reads are usually associated with DML (Data Manipulation Language) activity. Managing Oracle Database Systems: User and Reference Guide 210 Chapter 2—Using Foglight for Oracle DB Block Changes The number of database blocks in memory (SGA) that were modified by the session as a result of DML activity. Such changes generate redo log entries and then become permanent changes to the database if the transaction is committed. DB File Parallel Read Wait Time spent waiting for the completion of multiple input/output requests in parallel to read blocks from datafiles into memory. This wait event may occur during recovery or during regular activity when a session batches many single block input/output requests together and issues them in parallel. DB Link Wait The time spent on active wait events, excluding CPU wait events. This metric shows the ratio between time spent on IO wait events and the total time spent on wait events. Datafile Write Wait Time spent waiting for the DBWR (Database Writer) to perform a write to files and blocks (for example, flushing dirty data buffers from the SGA to the datafiles). DB File Parallel Read Wait Time spent waiting for the completion of multiple input/output requests in parallel to read blocks from datafiles into memory. This wait event may occur during recovery or during regular activity when a session batches many single block input/output requests together and issues them in parallel. Deferred Task Worker Wait Time spent by the various sessions waiting for I/O requests; this wait indicates that either a large number of suspect pages are encountered or the disk subsystem is overloaded. Degree of Parallelism The total number of server processes (master and slaves) assigned to handle a single operation. Direct Path IO Wait Time spent waiting for the completion of asynchronous input/output requests that bypass the buffer cache. Direct Path I/O usually reflects the disk sorts and bulk loads operations that require retrieving large amounts of data, which far exceed the buffer cache’s capacity. Managing Oracle Database Systems: User and Reference Guide 211 Chapter 2—Using Foglight for Oracle Disk Sorts The number of sort operations that require creation of temporary segments on disk to store the intermediate sort results. This occurs if the data being sorted cannot be fit into the memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. Disk Utilization The percentage of time the busiest disk spent serving system-wide I/O requests. This metric serves as a measure for the system I/O load. High values may indicate device bottleneck. Executions Ended The average number of statements whose activity completed during the specified time range. Executions Started The average number of statements that started running during the specified time range. External File IO Wait Time spent waiting for the completion of an input/output request on an external large object (LOB), such as a binary file (BFILE). External Procedures Wait Time spent by the various sessions waiting for external procedures to end. An external procedure, also sometimes referred to as an external routine, is a procedure stored in a DLL on Windows, or in shared library under UNIX. The external procedure is registered with the base language and then invoked to perform special-purpose processing. Because external procedures run in a process separated from the database instance, using these procedures ensures that any problems on the client side do not adversely affect the database. Free Buffer Wait This wait event occurs when the session needs a free buffer so it can bring a data block into the buffer cache, and is waiting for a buffer that is not dirty to become available. This can occur if the Database WRiter (DBWR) is not writing dirty buffers to disk fast enough. Full scan ratio The percentage of rows that were retrieved as a result of full table scans within the total retrieved rows. Full Scans Number of unrestricted full scans per second. These can be either base-table or full-index scans. Managing Oracle Database Systems: User and Reference Guide 212 Chapter 2—Using Foglight for Oracle Free Buffer Wait This wait event occurs when the session needs a free buffer, so it can bring a data block into the buffer cache, and is waiting for a buffer that is not dirty to become available. This can occur if DBWR is not writing dirty buffers to disk fast enough. IO Data Page Wait Time spent by the various sessions waiting to latch a buffer for an I/O request; this wait is related to issues with the disk I/O subsystem. IO Wait Time spent waiting for disk I/O operations to complete. I/O is one of the most expensive operations in a database system. SQL statements that are I/O intensive can monopolize memory and disk use and cause other database operations to compete for these resources. Generally, I/O Wait is caused by poorly tuned SQL queries or applications that generate a significant amount of logical I/O, which is translated into excessive physical disk usage. In this case, SQL/Application tuning can reduce the logical I/O-induced load. However, it could also be caused by poorly-configured disks or storage sub-systems. Latch Wait Time spent by the session being blocked by a latch, waiting for it to be released. Latches need not be locked for the duration of a transaction. They are low-overhead, short-term memory synchronization objects. They are used mostly to protect a row when it is queried for a connection. Latch Savepoint Wait Time spent by the various sessions waiting to synchronize commits to marked transactions. Lock Wait Time spent by the session being blocked, waiting for the blocking lock to be released. Lock Bulk Update Wait Time spent by the various sessions waiting to acquire bulk update locks. Bulk update locks are applied by the database engine when multiple threads perform bulk copying of data into the same table. Applying these locks allows the threads to bulk load data at the same time into the same table, while preventing access to the table from other processes that do not carry out data bulk loading. Lock Exclusive Wait Time spent by the various sessions waiting to acquire bulk update locks. Managing Oracle Database Systems: User and Reference Guide 213 Chapter 2—Using Foglight for Oracle Lock Intent Wait Time spent by the various sessions waiting to acquire intent locks. Intent locks are locks used by the database in order to protect a resource that is found lower in the lock hierarchy from a shared (S) lock or exclusive (X) lock. Because these locks are acquired before a lock at the lower level, they signal intent to place locks at a lower level. Lock Requests Number of new locks and lock conversions requested from the lock manager. Lock Schema Wait Time spent by the various sessions waiting to acquire schema locks. Schema locks are used for preventing changes to the underlying tables structures while the query is compiled. Lock Shared Wait Time spent by the various sessions waiting to acquire shared locks. Lock Update Wait Time spent by the various sessions waiting to acquire update locks. Lock Wait Time spent waiting for blocking locks to be released. Lock wait events occur when a session attempts to acquire a lock that is already held by another session. A session is blocked until the blocking session releases the lock. Locks are designed to ensure data integrity by limiting simultaneous data access. Multi-user database locking generally consists of two levels: exclusive locks and share locks. You want to watch out for exclusive locks (that is, TX) because they prohibit resource sharing. For example, the first transaction that exclusively locks a resource is the only one that can alter the resource (except for the DBA) until the exclusive lock is released. Share locks, unlike exclusive locks, allow a resource to be shared. Most locking issues are application-specific and can be addressed by tuning the concurrency logic in the application. Log Buffer Wait Time spent by the various sessions waiting for space in the log buffer or otherwise waiting for memory to be made available to write log records, while redo entries are being copied into the log buffer. Log buffer waits frequently occur when redo entries are written into the log buffer faster than LGWR can write these entries to the redo logs, thereby requiring user sessions to wait because of log write latency. Consistently high values of this metric can indicate the log devices’ inability to keep up with the amount of log generated by the server. Managing Oracle Database Systems: User and Reference Guide 214 Chapter 2—Using Foglight for Oracle Log Other Wait Time spent by the various sessions waiting for miscellaneous logs. Log Synchronization Wait Time spent by the various sessions waiting for log space to be freed by log truncation, that is, deleting log records from the transaction log. Even though log truncation is an automatic process, this process can be delayed as a result of several factors, such as long running transactions, a data backup in progress, or if no checkpoint occurred since the last log truncation. Log Switch Wait Time spent waiting for the completion of a redo log switch. A log switch is the operation succession that takes place when Oracle stops writing to the current log file (online redo log group), closes this file, opens the next log file, and begins writing to the new log file. This occurs automatically when the current online redo log file group becomes full. Log Switch Wait Ratio The percentage of time the session waited for the completion of redo log switches, out of the total time spent waiting by the instance. Under usual circumstances, log switches, which take place when LGWR stops writing to one online redo log group and starts writing to another, should not take long to complete, as they occur automatically when the current online redo log file group becomes full. Therefore, the portion of the log switch wait time within the instance’s total wait time should not exceed 5 percent. However, however log switch wait time can become critical under the following circumstances: • The next redo log to be used is not available because the checkpoint started before the completion of the log switch. • The redo log has not yet been archived. Log Wait The total amount of time that various processes spent waiting for the completion of a redo log operation. High values of this metric can indicate inefficient log process, resulting from reasons such as extensive log switches, failure to commit logs, or large log files whose archiving involves extended wait periods. Log Write Wait Time spent waiting for the completion of a write operation (physical input/output) to a log file. The Oracle log writer (LGWR) experiences log write wait events during the periodical flush of log buffer contents into the active log file. Managing Oracle Database Systems: User and Reference Guide 215 Chapter 2—Using Foglight for Oracle Memory Sorts The number of sort operations that were carried out completely in memory, thereby saving the need for disk writes. The sort operation can take place in memory if the sorted data can be fit into the memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. Memory Wait Time spent waiting by the various processes waiting for the completion of a Log operation. Miscellaneous Elapsed time spent waiting for the completion of miscellaneous operations, that is, operations that cannot be classified into any other wait categories. Miscellaneous IO Wait Time spent waiting during sporadic I/O activity, such as opening a datafile. Usually this type of wait event is negligible. Monitor Up Time The amount of time that the Foglight Performance Analysis agent was running during the current time range. Multi-Block Read wait Time spent waiting for multiple data blocks to be read from the datafiles into the SGA buffer. Multi-Block Read wait events usually reflect full table or index scans. Network HTTP Wait Time spent by the various sessions waiting for outstanding HTTP connections to complete and exit. Network IO Wait Time spent by the various sessions waiting for network packets. Network IPC Wait The Inter Process Communication (IPC) wait signifies time spent by the various sessions waiting for sub-tasks to generate data; long waits are indicative of unexpected blockages. Network Mirror Wait Time spent by the various sessions waiting for database mirroring events to complete. Network Wait Network wait events occur when a session spends time waiting for messages to be sent or received over the network interface. Managing Oracle Database Systems: User and Reference Guide 216 Chapter 2—Using Foglight for Oracle Network performance, which is measured in the number (per second) of packets sent and received, can be used just like disk statistics to detect overload or non-optimal performance within a network or a network interface. For more information, see “SQL *net” on page 206. Excessive network wait can result from either: • Excessive network usage, originating in the application • Physical issues, identifiable by network errors and network collisions Non-buffer Activities The time spent on active wait events, excluding buffer wait. Non-CPU Activities The time spent on active wait events, excluding CPU wait events. This metric shows the ratio between time spent on I/O wait events and the total time spent on wait events. Non IO Wait The time spent on active wait events, excluding I/O wait events. Note The Non I/O Wait metric is displayed only in the Breakdown view, for showing the ratio between time spent on I/O wait events and the total time spent on wait events. Non Latch Wait The time spent on active wait events, excluding latch wait. Non Lock Wait The time spent on active wait events, excluding lock wait. Non-other Wait The total time spent on active wait events that are distinctly classified, such as latch wait, buffer. Non Redo Wait The time spent on active wait events, excluding redo wait events. Oracle Data I/O Operations Total number of I/O operations (both physical and logical) performed by the Oracle session. Oracle Physical I/O Operations Total number of physical I/O operations (read, writes, and redo) performed by the monitored Oracle instance processes. Note A physical I/O operation issued by Oracle can be satisfied by the operating system's file cache. Managing Oracle Database Systems: User and Reference Guide 217 Chapter 2—Using Foglight for Oracle Other DB Buffer Wait Sporadic buffer waits caused by buffer contention, such as buffer write wait; usually negligible. Other CPU Usage CPU consumption of processes not related to Oracle or its associated services. Other Miscellaneous Wait Time spent by the various sessions waiting for miscellaneous database operations. Other Wait Time spent by the various sessions waiting for other database operations. Overall CPU The amount of time Oracle spent both waiting for CPU and using CPU during the specified time range. Page Life Expectancy Number of seconds a page will stay in the buffer pool without references. Low values of this metric require the buffer pool to frequently flush the page to the physical disk, thereby possibly indicating a memory issue. Such an issue should be checked to see if it results from lack of physical memory, inefficient queries, and other related conditions. Page Splits Number of page splits taking place. Page splitting occurs when an index or data page becomes full and then is split between the existing page and a newly allocated page. Excess page splitting can lead to I/O performance issues, indicating a need to rebuild indexes and/ or increase the indexes’ fill factor. Paging Total number of page operations (Page-in and Page-outs) performed by the operating system. Physical I/O The amount of physical I/O operations (both read and write) carried out by Oracle during the specified time range. Physical Memory Used The amount of physical memory, or RAM, being used by the server. Physical Page Reads The number of physical database page reads issued during the specified time range. Managing Oracle Database Systems: User and Reference Guide 218 Chapter 2—Using Foglight for Oracle Physical Page Writes The number of physical database page writes issued during the specified time range. Physical Reads The number of data blocks read by Oracle, which are not present in the buffer cache or in the PGA (Program Global Area) memory, resulting in I/O read operations. This metric indicates the total number of data blocks that are read from disk, plotted over time. Physical Writes The total number of data blocks written to disk. This metric represents the rate at which Oracle generates system I/O calls, plotted over time. Recursive Calls Recursive calls, also known as recursive SQL statements, are additional, internal SQL statements that Oracle occasional executes as part of executing an SQL statement issued by a user. Such calls modify the tables maintained for internal processing. Redo Log Write The total number of write operations to the redo log files, plotted over time. Redo Wait The elapsed time including the contention resulting from: • User sessions writing redo entries to the redo log buffer • LGWR writing the log buffer contents into the active redo log file • ARCH archiving redo logs to the destination location Redo Writes The total number of write operations issued by LGWR to the redo log files. Resource Manager Wait Wait events that sessions experience when they exceed thresholds/quotas that the resource manager permits them. Resource Manager (Oracle 10g and later) can be used to provide more fine-grained control of resource allocations to services in both RAC and single instance environments. These two facilities are integrated so that services can be mapped to Resource Manager consumer groups. Row Source Ratio The percentage of the total rows retrieved using full table scans within the total number of retrieved rows. Significant values may indicate non-optimal queries. Managing Oracle Database Systems: User and Reference Guide 219 Chapter 2—Using Foglight for Oracle Samples Number of SGA samples taken during the specified time range. Service Broker Wait Time spent by the various sessions waiting for Service Broker event handlers and endpoints. Session Logical Reads The number of data blocks read by Oracle, which were already present in the buffer cache, thereby saving I/O read operations. This value is the total of Current Reads (also known as Database Block Gets) plus Consistent Reads (also known as Consistent Gets). For details, see “DB Block Gets” on page 209 and “Consistent Gets” on page 209. Session Logons The number of sessions that logged on during the current interval. Session Logoffs The number of sessions that logged off during the current interval. Single Block Read Wait Time spent waiting for a single block to be read from a datafile into the SGA buffer. Sort Overflow Ratio Measures the portion of sorts that could not fit into a session's sort area and therefore used temporary segments. High values may indicate a need to increase the value of SORT_AREA_SIZE parameter (in the initialization files), in order to allow more memory space for sort operations. SQL Executions The number of SQL statements executed during the specified time range. SQL Recompilations The number of SQL recompilations during the specified time range. Synchronous Task Wait Time spent by the various sessions waiting for tasks that started synchronously. Table Fetch by Row ID The number of rows that are fetched using a row ID (usually recovered from an index); each row returned increments this value. Managing Oracle Database Systems: User and Reference Guide 220 Chapter 2—Using Foglight for Oracle Table Fetch Continued Row The number of times a chained or migrated row that spans more than one block is encountered during a fetch; each chained or migrated row encountered increments this value. Table Lock Escalation Lock escalation takes place when a specific query, which obtains a large number of locks at the row or page level, is automatically upgraded to a table lock level. This upgrade is being carried out because granting a single lock at this level is deemed more effective than creating and granting many locks at lower levels. Table Scan Blocks Gotten Occurs during scanning operations, when each row is retrieved sequentially by Oracle. Each block encountered during the scan increments this value. Table Scan Rows Gotten The number of rows scanned during full table scans. During scanning operations, each row is retrieved sequentially by Oracle. This statistic counts the number of rows encountered during the scan block operation. High values may indicate nonoptimal queries, the absence of proper indexing, or both. Table Scans Short Tables Number of times Oracle has completely scanned a short table. A table is defined as short if: • The table's high water mark is less than five blocks or 2% of the database block buffer cache; or: • The table's high water mark is less than the value of CACHE_SIZE_THRESHOLD (v8.0 or earlier); or: • The table is defined with the CACHE option set (v8i and later) Target Instance Memory The total amount of dynamic memory the Oracle instance is willing to consume. If the value of this metric is significantly higher than that of the Total SQL Instance Memory metric, it can indicate that the system is low on physical memory. Total CPU Usage Overall CPU utilization by all processes running on the host (both OS and Oracle processes). Total IO Wait Time spent waiting for disk I/O operations to complete. Managing Oracle Database Systems: User and Reference Guide 221 Chapter 2—Using Foglight for Oracle I/O is one of the most expensive operations in a database system. I/O-intensive SQL statements can monopolize memory and disk use and cause other database operations to compete for these resources. Generally, I/O Wait is caused by poorly-tuned SQL queries or applications that generate a significant amount of logical I/O, which translates into excessive physical disk usage. In this case, SQL/application tuning can reduce the logical I/O- induced load. However, it could also be caused by poorly-configured disks or storage sub-systems. Note The Total IO Wait metric is displayed only in the Baseline view. Total Redo Wait The elapsed time including the contention resulting from: • User sessions writing redo entries to the redo log buffer • LGWR writing the log buffer contents into the active redo log file • ARCH archiving redo logs to the destination location Total Sorts The summary of database disk sorts and memory sorts. For details, see “Disk Sorts” on page 211 and “Memory Sorts” on page 215. User Calls Number of user calls such as login, parse, fetch, or execute. This metric value represents the rate at which the user application addresses Oracle. User Calls per Parse This rate indicates how well the application is managing its context areas. It measures the average amount of calls in relation to statement parses. Changes in this rate are most likely to arise from application change, but may also indicate that usage patterns are changing and users are moving from one module to another, either more frequently or less frequently. An increasing value of this metric indicates that the effectiveness of the shared pool has improved, resulting in lesser need for parses. The ratio is calculated as follows: User Calls/Parse Count 3 Reference This chapter describes the rules and collections that are used with monitored Oracle databases. • Foglight for Oracle Rules • Overview of Collections Foglight for Oracle Rules Foglight for Oracle includes a set of predefined rules that are applied to the incoming monitoring data. The rules that monitor SQL Server instances and databases have the prefix DBO. You can and should modify rules to better suit your environment. To modify Foglight for Oracle rule threshold values, alarm settings, and alarm email notifications, use the Databases Administration — Alarms View. To review all Foglight rules, copy rules, edit rules, or create new user-defined rules, use the Foglight Administration — Rules Management Dashboard. Tip For rules that reference registry variables for threshold values, modify the threshold in the registry variable, rather than modifying the rule. For help finding and editing registry variables, search for “Registry Variable” in the online help. Databases Administration — Alarms View In the Alarms view, you can customize alarm settings and email notifications for all alarms, a category of rules, or a single rule. The threshold changes can be localized to one agent or all selected agents. Note When you make edits to alarm settings on the Alarms view, the edits are saved separately and applied over the predefined rules, which protects your changes from software updates. To manage alarm thresholds and notifications for rules: 1 In the navigation panel, under Homes, click Databases. 2 Select the row check boxes beside one or more Foglight for Oracle instances. 3 Click Agent settings and then click Administration. The Databases Administration dashboard opens. Managing Oracle Database Systems: User and Reference Guide 223 Chapter 3—Reference 4 Click Alarms. The Alarms view contains the predefined rules organized by category. 5 From here, you can perform the following tasks: • • • • • Set alarm sensitivity levels Enable/disable alarms Change alarm threshold values Enable/disable severity levels for each alarm Configure email notifications For help with these tasks, see “Customizing Alarms for Foglight for Oracle Rules” on page 175. Foglight Administration — Rules Management Dashboard In the Rules Management dashboard, all Foglight for Oracle rules are displayed in the list. You can restrict the list to show only the rules installed with a cartridge. You can also search the list for a single rule or a group of rules. Note Predefined rules may be modified during regular software updates. To avoid losing your changes to these rules, we recommend copying the rule and making edits to the copy. Enable the copy and disable the original rule. You may want to identify your custom rules with a unique prefix. To review and edit rules: 1 In the navigation panel, under Homes, click Administration. 2 In the Administration dashboard, click Rules. The Rule Management dashboard opens. 3 To show only Foglight for Oracle rules, in the Search box, type the prefix: DBO Tip Alternatively, you can select DB_Oracle_UI from the Cartridge list. 4 From here, you can perform the following tasks: • Review a short description of the rule. • Review and edit threshold values. Note For DBO rules, use the Databases Administration — Alarms View instead. • Copy rules. • Edit rule conditions. • Associate actions with rules. Note By default, predefined Foglight for Oracle rules do not trigger any actions. • Create user-defined rules. Note User-defined rules are always managed from the Rule Management dashboard; these rules are not displayed in the Databases Administration — Alarms View. For help with these tasks, open the online help from the Rules Management dashboard. Managing Oracle Database Systems: User and Reference Guide 224 Chapter 3—Reference Overview of Collections Foglight for Oracle collects raw data for a set of collection types. This data is used by the Rules to trigger alarms and is used to populate dashboards. The collections data tables are not viewable from Foglight for Oracle, but the data is implemented in the various drilldowns of Foglight for Oracle. The collection data tables are not viewable from Foglight for Oracle. Note Several collections contain metrics whose name ends with Rate (for example: DBO Wait Buffer Busy Rate). Such metrics, which are not documented in this guide, are used for plotting the original metric (usually called by the same name without the “Rate” suffix) on a graph for the selected time range. Advisories Purpose The Advisories collection collects the summary of the Performance Analysis advisories. Collection Type Performance Analysis Collection Sampling Settings Frequency Mode Collection Interval (seconds) Realtime Collection 900 Online 900 Offline 900 Metric Descriptions Metric Description Additional info Additional information about the advisory. Advisory Name The advisory’s name. Linked PA View The resource to which the link included in the advisory points. Message The message displayed by the advisory. Occurrence End Time The time when the occurrence reported by the advisory ended. Managing Oracle Database Systems: User and Reference Guide 225 Chapter 3—Reference Metric Description Occurrence Start Time The time when the occurrence reported by the advisory started. Severity The advisory’s severity. Stealth Collect Host The name of the host where the StealthCollect agent is installed. Stealth Collect Instance The name of the StealthCollect instance. Stealth Collect Port The number of the StealthCollect port. Type The advisory’s type. Alert Log Purpose The Alert Log collection retrieves the content of Oracle alert log. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 900 Offline 900 Metric Descriptions Metric Description Critical Errors The number of critical error notifications that were received during the specified time range. Dump File Name Indicates the name of dump file (if this name appears in the alert log message). Fatal Errors The number of fatal error notifications that were received during the specified time range. Informational Errors The number of informational messages that were received during the specified time range. Managing Oracle Database Systems: User and Reference Guide 226 Chapter 3—Reference Metric Description Message The text of the alert log message. Name The name of the alert log message. Severity Indicates the alert log message severity (Fatal, Critical, Warning, Informational). Timestamp Alert log message timestamp. Type Alert log message type (Internal Error, Storage, Media Failure and so on). Warning Errors The number of warning error notifications that were received during the specified time range. Archive Destination Miscellaneous Purpose The Archive Destination Miscellaneous collection provides Archive Destinations activity rates and time considerations. Collection Type Oracle and Operating System Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 3600 Online 3600 Offline 3600 Metric Descriptions x‘ Metric Description Archive Destination Free Space The free space in the file system where the most critical archive destination is located. Archive Destination Size The size of the file system where the most critical archive destination is located. Archive Rate The rate of the archiving operation. Managing Oracle Database Systems: User and Reference Guide 227 Chapter 3—Reference Metric Description Archiving Volume Archiving volume Average Archive File Size The size of the average archive file Binding Binding type: • MANDATORY — successful archival is required. • OPTIONAL — successful archival is not required. Critical Archive Time To Failure The number of hours left until the critical archive destination fills. Database The database name. Destination The destination ID of the most critical archive destination. Destination directory The archive’s destination directory Destination Free Space The total amount (in megabytes) of free space available for the archive destination target. Destination Total The total size (in megabytes) of the archive destination target. Time To Failure The number of hours remaining until the archive destination with the shortest time to failure is filled. Archive Logs Storage Purpose The Archive Logs Storage collection provides Archive Destinations properties and file systems space data. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 900 Offline 3600 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Archive Destination The name of the destination archive file. Archiver Status The archiver status; can have one of the following values: Stopped, Started, and Failed. Binding Binding type: • MANDATORY — successful archival is required. • OPTIONAL — successful archival is not required. Database Database name Destination Directory The directory used for storing the archive. Destination End The value of destination directory is replaced by Recovery Destination if the directory is used for recovery. Destination Free Space The total amount (in megabytes) of free space available for the archive destination target. Destination Total The total size (in megabytes) of the archive destination target. Destination Type The type of the archived log destination: • PUBLIC • PRIVATE Destination Usable The amount (in megabytes) of storage space usable for the archive logs in the destination directory. Error Displays the error text. If database in archive mode Indicates whether the database is in archive mode Is ASM Indicates whether the instance is an Automatic Storage Management (ASM) instance Is Recovery Indicates whether the directory is used for recovery (True) or not (False). Log Sequence Identifies the sequence number of the most recently archived redo log. Multiple destinations in filesystem Indicates if more than one archive destinations reside on the same filesystem. Re-open Identifies the retry time (in seconds) to open a destination after error. 228 Managing Oracle Database Systems: User and Reference Guide 229 Chapter 3—Reference Metric Description Space Reclaimable The amount of space that can be used if files are removed from the Flash Recovery Area. Status The current status of the destination in V$ARCHIVE_DEST: • VALID • INACTIVE • DEFERRED • ERROR • DISABLED • BAD PARAM • ALTERNATE • FULL Target Specifies whether the archive destination is local or remote to the primary database: • PRIMARY - local • STANDBY - remote Transmit Mode The mode in which data is transmitted across the network: • SYNCHRONOUS • ASYNCHRONOUS • PARALLELSYNC Unarchived redo logs The percentage of redo logs files that are waiting for archiving Valid Type The redo log type or types that are valid for the destination: • ONLINE_LOGFILE • STANDBY_LOGFILE • ALL_LOGFILES ASM Diskgroups Purpose The ASM Diskgroups collection collects storage and activity data regarding ASM Diskgroups. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Frequency Mode Collection Interval (Seconds) Online 300 Offline 900 Metric Descriptions Metric Description Allocation Size Unit The size of allocation units in the disk group. ASM Metadata Block Size The block size for ASM metadata in the disk group. Average Service Time The average service time for all disks in the group. Group Name The name of the disk group. Group Number The number assigned to an ASM disk group. Group State The state of the specified ASM disk group relative to the Oracle instance (CONNECTED, BROKEN, UNKNOWN, DISMOUNTED, or MOUNTED). Number of Disks The number of disks in the disk group. Offline Disks The number of disks in the disk group that are offline. Note This information is unavailable for Oracle 10.1 Read Errors The total number of failed I/O read requests for the disk group. Read IO Requests The total number of I/O read requests for the disk group. Total Group Size The total capacity of the specified ASM disk group (in megabytes). Usable Group Space The unused space on the disk group (in megabytes). Note This information is unavailable for Oracle 10.1 Used Group Space The percentage of used disk space on the entire disk group. This value is calculated as follows: 100*(Total Group Size - Usable Group Space) Total Group Size Write Errors The total number of failed I/O write requests for the disk group. 230 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Write IO Requests The total number of I/O write requests for the disk group. ASM Diskgroups Balance Purpose The ASM Diskgroups Balance collection estimates the rate of balance of IO Activity and Response time of ASM disks on ASM diskgroups. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 300 Offline 900 Metric Descriptions Metric Description Balanced IO The extent, in percents, to which the I/O activity is balanced across the ASM diskgroups. Note Balanced Time The extent, in percents, to which the I/O time is balanced across the ASM diskgroups. Note Group Number This information is unavailable for Oracle 10.1 This information is unavailable for Oracle 10.1 The number assigned to an ASM disk group. ASM Disks Purpose The ASM Disks collection collects storage and activity data regarding ASM disks. Collection Type Oracle 231 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 300 Offline 900 Metric Descriptions Metric Description Average Service Time The service time for each disk. Service time is the response time of a disk, plus any wait time experienced during disk operations. Create Date The date and time when the disk was added to the disk group. Disk Free Space The unused capacity of the ASM disk. Disk Name The name of the disk. Disk Number The number assigned to an ASM disk within its disk group. Disk Path The disk’s path, as indicated on the operating-system. Disk Size The total capacity of the ASM disk. Disk State The state of the disk, regarding its ASM disk group (UNKNOWN, NORMAL, ADDING, DROPPING, HUNG, FORCING, or DROPPED). Fail Group The name of the failure group that contains the disk. Group Number The number assigned to an ASM disk group. Header Status The status of the disk at the discovery stage (UNKNOWN, CANDIDATE, INCOMPATIBLE, PROVISIONED, MEMBER, FORMER, CONFLICT, or FOREIGN). Label The label portion of the disk name at the discovery stage. Mode Status The access type granted to the disk (UNKNOWN, ONLINE, or OFFLINE). Read/write access is only granted to ONLINE disks; OFFLINE disks still remain part of their disk group. Mount Date The date and time when the disk was first mounted. Mount Status The mount status of the disk (MISSING, CLOSED, OPENED, CACHED, IGNORED, or CLOSING). 232 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Read Errors The total number of failed I/O read requests for the disk. Such failures can result from a loss of access to the entire disk, or media corruptions on a generally healthy disk. Read Time The total time spent on I/O read requests. Redundancy The external redundancy of the disk, if the ASM mechanism does not provide redundancy. The possible states are: UNKNOWN, UNPROT, MIRROR, or PARITY. Total IO Percent The number of I/O operations performed by the selected disk as a percentage of all I/O performed by all disks in all ASM disk groups assigned to the Oracle instance. Total Reads The total number of I/O read requests for the disk. Total Size Percent The size of the disk as a percentage of the size of all disks in all ASM disk groups assigned to the Oracle instance. Total Writes The total number of I/O write requests for the disk. Used Space Pct The percent of used space out of the entire space allocation on the ASM disk. Write Errors The total number of failed I/O write requests for the disk. Write Time The total time spent on I/O write requests. ASM Instance Total Purpose The ASM Instance Total collection provides ASM-related data, summarized for the database instance. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 300 Offline 900 233 Managing Oracle Database Systems: User and Reference Guide 234 Chapter 3—Reference Metric Descriptions Metric Description Total Disk Space Allocated The total capacity allocated to all ASM disks residing on ASM groups that are used by this database instance. Total Disks Free Percent The total free space percentage on all ASM disks residing on ASM groups that are used by this database instance. Total IO The total number of I/O operations (reads and writes) on ASM managed disks. Total Reads The total number of I/O read requests for ASM managed disks. Total Writes The total number of I/O writes to ASM managed disks. ASM Operations Purpose The ASM Operations collection retrieves details regarding the currently running ASM operations. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 300 Offline 900 Metric Descriptions Metric Description Actual Power The power actually allocated to the operation. Estimated Minutes The estimated amount of time (in minutes) that the operation is expected to take. Estimated Rate The estimated rate (%) at which allocation units will be moved by the operation Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Estimated Work The estimated number of allocation units that the operation needs to move. Group Number The number of the disk group where the operation is running. Operation The type of the operation; REBAL, COD, DSCV, ACD, RFRSH, or EXPEL. Operation State The state of the operation; WAIT, RUN, REAP, HALT, or ERRORS. Requested Power The power requested for the operation as specified by the ASM_POWER_LIMIT parameter. Higher value of ASM_POWER_LIMIT indicate faster operation, while lower values indicate slower operation that uses fewer resources. So Far Moved Units Number The number of allocation units moved so far by the operation. Backup Config Purpose The Backup Config collection provides a list of configuration parameters value for RMAN. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 21600 Online 21600 Offline 21600 Metric Descriptions Metric Description Configuration # The number of configuration parameters 235 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Configuration Name The names of the configuration parameters Configuration Value The values of the configuration parameters Database Database name Backup Jobs Purpose The Backup Jobs collection collects RMAN jobs statuses. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 21600 Online 21600 Offline 21600 Metric Descriptions Metric Description Backup Job ID The ID of the backup job Backup Job Status The status of the backup job Backup Job Type The type of the backup job Database Database name Job End Time The end time of the backup job Job Start Time The start time of the backup job Output The output message of the backup job Session Rec ID The Oracle identifier of the RMAN job Session Stamp The time stamp of the session 236 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Backup Type Status Purpose Backup Type Status collection collects the last time stamp of each RMAN job that ran successfully. Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 21600 Online 21600 Offline 21600 Metric Descriptions Metric Description Backup Type The type of the backup job. Days Since Backup The number of days that passed since the last database backup. Last Successful Backup The date of the most recent successful backup that has been carried out for the database. Change Tracking Purpose The Change Tracking collection provides change tracking data, which is only available for a cartridge with the Performance Analysis agent installed and configured. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 3600 Online 3600 Offline 3600 237 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Descriptions Metric Description Additional Info Additional information about the change, if applicable. Change Type (change, new, delete) The type of the change: modification, deletion or addition of a new item Changes in Oracle Schema The changes that took place in the Oracle schema Count The number of changes of the selected type. Occurrence End Time The time when the change tracking ended Occurrence Start Time The time when the change tracking started Stealth Collect Host The name of the Foglight Performance Analysis host Stealth Collect Instance The name of the Foglight Performance Analysis instance Stealth Collect Port The name of the port used for communicating with Foglight Performance Analysis Change Date The date when the change in the execution plan took place. SQL Hash Value The hash value of the selected SQL statement. SQL Text The text of the selected SQL statement. Cluster Latency Purpose The Cluster Latency collection provides Cluster Latency and Activity data. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 238 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Descriptions Metric Description Average Global Cache Convert Time Average amount of time required for the completion of global conversions on GCS resources. Average Global Lock Convert Time Average amount of time required for the completion of synchronous and asynchronous global lock converts. Average Global Lock Get Time Average amount of time required for the completion of synchronous and asynchronous global lock gets. Balance Interconnect Load balance between the various cluster nodes. Block Received Latency The average transfer time of a block from another instance. Block Sent Latency The average time needed to prepare a requested block. Blocks Corrupt Percent The percentage of global cache blocks that encountered a corruption or checksum failure within the entire number of global cache blocks sent over the Interconnect. Blocks from Interconnect The number of blocks received from Interconnect after the requests for blocks were not satisfied from the local instance, and therefore had to make a request for the block from the remote. Blocks to Interconnect The number of blocks sent via Interconnect after the requests for blocks were not satisfied on the requesting instance, and therefore had to make a request for the block from the local instance. Blocks/sec from Interconnect The rate per second of blocks received from Interconnect after the requests for blocks were not satisfied from the local instance, and therefore had to make a request for the block from the remote. Blocks/sec to Interconnect The rate per second of blocks sent via Interconnect after the requests for blocks were not satisfied on the requesting instance, and therefore had to make a request for the block from the local instance. Cluster Latency Rate The average time it took to transfer a block through the Interconnect, that is, the time that has elapsed since the request for the block was submitted until the block was received by the requesting instance. Consistent Gets The number of times a consistent read was requested for a block by this session 239 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description CR Blocks Send Time Total amount of time spent sending constructed CR Blocks through the Interconnect. Current Block Send Time Total amount of time spent sending Current Blocks through the Interconnect. DB Block Gets The number of data block reads in CURRENT mode (namely, current reads). GC Blocks Corrupt The total number of global cache blocks that encountered a corruption or checksum failure during the specified time range As such failures occur during data transfer, high values of this metric can indicate network-related issues. GC Blocks Lost Total number of global cache blocks lost during the specified time range. As the loss of global cache blocks takes place during data transfer, high values of this metric can indicate network-related issues. GC CR Block Build Time The average time taken by an LMS process to create a consistent read block on a holding instance within the cluster. GC CR Block Flush Time The average time taken by the LGWR (log writer) process to flush any changes required to a buffer on the log flush queue. GC CR Block Receive Time Total amount of time foreground processes spent waiting for a CR block to be sent through the Interconnect. GC CR Block Send Time The average time taken by an LMS process to initiate the sending of a consistent read block. This is NOT the same as the time elapsed before the block arrives at the requesting instance. GC CR Blocks Received Total number of consistent read data blocks that Oracle processes did not find in their local cache and had to fetch from the global cache, using a consistent read request. GC CR Blocks Served Total number of blocks constructed by the Block Server Process (BSP). GC Current Block Flush Time The average time taken by the LGWR (log writer) process to flush the changes from a block to disk (a forced log flush) before the block is shipped to the requesting instance. GC Current Block Pin Time The average time taken by an LMS process to pin the current block before shipping it to the requesting instance. GC Current Block Receive Time Total amount of time foreground processes waited for a current block to be sent through the Interconnect. 240 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description GC Current Block Send Time The average time taken by an LMS process to initiate the sending of a current block. Note This metric is not identical to the time elapsed before the block arrives at the requesting instance. GC Current Blocks Received Total amount of time foreground processes waited for a current block to be sent through the Interconnect. GC Current Blocks Served The number of current blocks shipped to the requesting instance over the hardware Interconnect. Global Cache Convert Time Total amount of time spent waiting for global cache converts on all sessions. Global Cache Converts Resource converts of buffer cache blocks. This statistic is incremented whenever GCS resources are converted from Null to Exclusive, Shared to Exclusive, or Null to Shared. Global Lock Convert Time Total time spent waiting for all synchronous and asynchronous global lock converts. Global Lock Get Time Total elapsed time of all synchronous and asynchronous global lock gets. Global Lock Releases Total number of synchronous global lock releases. Global Lock Sync Converts The total number of resources that Oracle converted from an incompatible mode during the selected time range. Interconnect Block Latency The average time it takes to prepare and transfer a block across the Interconnect network. Lost Blocks Ratio The ratio of global cache blocks losses during data transfer via the Interconnect. Received from Interconnect Time Total amount of time the foreground processes waited for blocks to be sent through the Interconnect Sent to Interconnect Time Total amount of time spent sending blocks through the Interconnect Total CR Blocks The total number of CR blocks that were served/received through the Interconnect Total Current Blocks The total number of current blocks that were served/received through the Interconnect. 241 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Total I/C Load Total number of blocks (both CR and current) that were served and received from the Interconnect. Cluster Misses Purpose The Cluster Misses collection collects data regarding global cache misses. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval Realtime Collection 20 Online 60 Offline 300 Metric Descriptions Metric Description 2-way Misses The number of current and consistent blocks that were received by the selected instance from a specific instance, after a 2-way roundtrip. 3-way Misses The number of current and consistent blocks that were received by the selected instance from a specific instance, after a 3-way roundtrip. Consistent Blocks 2-way Misses The number of consistent blocks that were received by the selected instance from a specific instance, after a 2-way roundtrip. Consistent Blocks 3-way Misses The number of consistent blocks that were received by the selected instance from a specific instance, after a 3-way roundtrip. Consistent Blocks Misses The number of consistent blocks that were received by the selected instance from a specific instance, after a 2-way or 3way roundtrip. 242 Managing Oracle Database Systems: User and Reference Guide 243 Chapter 3—Reference Metric Description Current Blocks 2-way Misses The number of current blocks that were received by the selected instance from a specific instance, after a 2-way roundtrip. Current Blocks 3-way Misses .The number of current blocks that were received by the selected instance from a specific instance, after a 3-way roundtrip Current Blocks Misses The number of current blocks that were received by the selected instance from a specific instance, after a 2-way or 3way roundtrip. Datafile Fragmentation Purpose The Datafile Fragmentation collection collects information regarding the layout of the free space left in the datafile due to fragmentation. Collection Type Oracle Collection Sampling Settings None (on-demand collection). Metric Descriptions Metric Description Datafile name The name of the datafile Extent Size Offset end of the extent from the beginning of the file. Extent Start Offset start of the extent from the beginning of the file. Extent Type Defines the type of the extent: Used, Fully Usable, Partly Usable or Unusable. Sequence ID The datafile’s sequence ID. Datafile IO Activity Purpose The Datafile IO Activity collection collects metrics and performance ratios related to datafile I/O activity. Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Offline 900 Metric Descriptions Metric Description Avg IO Time Average time spent on a single IO request for a specific datafile Avg Read Time Average time spent on a single physical read request for a specific datafile Avg Write Time Average time spent on a single physical write request for a specific datafile Contents The type of the tablespace contents (temporary, permanent, undo) File Full Name The name of the datafile (including full path) File ID The datafile ID File Name The name of the datafile (without the path) File Path The datafile path IO Time The total time spent on carrying out IO operations (reading and writing data) on a specific datafile Last IO Time The time spent while performing last I/O operation on a specific datafile Max Single Read Time The maximum time spent while reading a single block from a specific datafile Max Single Write Time The maximum time spent while writing a single block to a specific datafile Min Single IO Time The minimum time spent while reading a single block from a specific datafile 244 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description PDB In multitenant architectures, the name of the pluggable database. Physical IO Total size, in blocks, of all disk reads and writes for a specific datafile Physical IO Percent The IO activity (block reads + block writes) for a specific datafile as a percentage of the total IO activity for all datafiles Physical Reads The number of data blocks read by Oracle, which are not present in the buffer cache or in the PGA, thereby requiring I/O read operations, that is: the total number of data blocks read from disk. Physical Reads Percent The block reads carried out for a specific datafile as a percentage of the of total block reads carried out for all datafiles Physical Writes Total size in blocks of all disk writes for a specific datafile Physical Writes Percent The block writes carried out for a specific datafile as a percentage of the of total block writes carried out for all datafiles Read Requests The number of read requests for a specific datafile that read one or more database blocks per request. Read Time The total time spent on reading data from a specific datafile. Single Block Read Time The total time spent on read requests that read one database block per request from a specific datafile. Single Block Reads Number of read requests for a specific datafile, which read one database block per request. Tablespace Name The tablespace name. Write Requests The number of write requests for a specific datafile that wrote one or more database blocks per request. Write Time The total time spent on writing data to specific datafile. Datafiles Storage Purpose The Datafiles Storage collection provides a list of all datafiles and their overall properties and storage details. 245 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Collection Type Oracle and Operating System Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 900 Online 900 Offline 3600 Metric Descriptions Metric Description Autoextend Indicates whether the datafile is autoextensible. Autoextensible Indicates if at least one datafile in the database is not autoextensible File ID The ID of the selected datafile. File Name The name of the datafile. Filesystem Name The name of the file system where the datafile resides. Free The amount of free space on the selected datafile. Free Pct The percentage of free space on the selected datafile. Is ASM File Indicates whether the datafile is an Automatic Storage Management (ASM) file. Maximum Used Percent The maximum percent of the used disk space of all the database datafiles. Maxsize The maximum size of this datafile; used for autoextensible datafiles. PDB In multitenant architectures, the name of the pluggable database. Size The size of the datafile (in megabytes). Status Indicates whether the datafile is available or invalid. Tablespace Name The name of the table space to which the datafile belongs. 246 Managing Oracle Database Systems: User and Reference Guide 247 Chapter 3—Reference Metric Description Total Allocated Space The total allocated space in the database. Total Allocated Space Delta Total Free Space Percent The percentage of free space in the database datafiles. Used The size (in megabytes) of used datafile storage. Used Pct The percentage of used datafile storage. Dataguard Destination Status Purpose The Dataguard Destination Status collection retrieves status and statistics for standby destinations in Data Guard. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 300 Offline 300 Metric Descriptions Metric Description Destination Directory The dataguard destination directory Destination Error Displays the error message for destination status. Destination ID The destination ID Managing Oracle Database Systems: User and Reference Guide 248 Chapter 3—Reference Metric Description Destination Status Identifies the current status of the destination: • VALID — initialized and available • INACTIVE — no destination information • DEFERRED — manually disabled by the user • ERROR — error during open or copy • DISABLED — disabled after error • BAD PARAM — parameter has errors • ALTERNATE — destination is in an alternate state • FULL — exceeded quota size for the destination Gap in Applied Logs The gap between the last archive log on the primary database and last applied archive log on the standby database. Gap in Received Logs The gap between the last archive log on the primary database and the last received archive log on standby database. Last Applied Sequence# Number of the last applied sequence Last Dataguard error The most recent error reported by Oracle for this standby destination Last Dataguard message The most recent DataGuard message reported by Oracle for this standby destination. Last Received Sequence# Number of the last received sequence. Dataguard Redo Apply Purpose The Dataguard Redo Apply Rates collection retrieves performance data for database configured with Data Guard. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 3600 Online 3600 Offline 3600 Managing Oracle Database Systems: User and Reference Guide 249 Chapter 3—Reference Metric Descriptions Metric Description Destination ID The destination ID Rate of Redo Apply The amount of MB of the redo logs applied to the standby database. Rate of Redo Receive The amount of MB of the redo logs received on the standby database. Dataguard Status Purpose The Dataguard Status collection retrieves general definitions and status of database configured with Data Guard. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 300 Offline 300 Metric Descriptions Metric Description Database Role The database role Number of Logs not Applied on Standby Number of logs that were not applied on standby mode. Number of Logs not Received on Standby Number of logs that were not received on standby mode. Number of Online Redo Logs The number of Online redo logs defined for the standby database. Number of Standby Redo Logs The number of Standby redo logs defined for the standby database. Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Protection Level Aggregated protection mode currently in effect for the database: • MAXIMUM PROTECTION — Database is running in maximized protection mode • MAXIMUM AVAILABILITY — Database is running in maximized availability mode • RESYNCHRONIZATION — Database is running in resynchronization mode • MAXIMUM PERFORMANCE — Database is running in maximized protection mode • UNPROTECTED — Database is unprotected (this normally occurs when the primary database is mounted and not open) Protection Mode Protection mode currently in effect for the database: • MAXIMUM PROTECTION — Database is running in maximized protection mode • MAXIMUM AVAILABILITY — Database is running in maximized availability mode • RESYNCHRONIZATION — Database is running in resynchronization mode • MAXIMUM PERFORMANCE — Database is running in maximized protection mode • UNPROTECTED — Database is unprotected (this normally occurs when the primary database is mounted and not open) Redo Logs Configuration Mismatch The number of online redo logs and the number of standby redo logs on the primary database do no match. Total Number of Dataguard Errors The number of DataGuard errors reported by Oracle for the standby database. Dump File Purpose The Dump File collection retrieves the contents of the Oracle dump file. Collection Type Oracle Collection Sampling Settings None (on-demand collection). 250 Managing Oracle Database Systems: User and Reference Guide 251 Chapter 3—Reference Metric Descriptions Metric Description Dump File Name The name of the dump file Message The dump file message Event Wait Purpose The Event Wait collection retrieves the instance wait events (workload) that took place within the specified time range. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions Metric Description Event ID The wait event’s ID Event Name The wait event’s name PDB In multitenant architectures, the name of the pluggable database. Wait Time The elapsed time of the selected wait event, plotted over time Exadata Cell Flashcache Purpose The Exadata Cell Flashcache collection retrieves performance, utilization and usage statistics about the Exadata cell flashcache. Managing Oracle Database Systems: User and Reference Guide 252 Chapter 3—Reference Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Offline 900 Metric Descriptions Metric Description Cell ID The Cell ID, provided by the hardware vendor. Flashcache hit ratio The ratio, within the total physical disk requests, of physical disks requests that were resolved from the Flashcache, thereby saving the need to carry out physical disk reads. Flashcache I/O Errors The number of IO errors that were encountered on the Flashcache Flashcache Keep read request misses (Keep Objects) The number of read IO requests for 'keep' objects that did not find all data in Flashcache and therefore had to perform physical read operations. Flashcache Keep read requests (Total) The number of read IO requests for 'keep' objects that were resolved from the Flashcache, thereby saving the need to carry out physical disk reads. Flashcache read request misses (Total) The total number of read IO requests that did not find all the requested data in the Flashcache, and therefore had to perform read from disks. Flashcache read requests (Keep objects) The number of read IO requests that were resolved from the Flashcache, thereby saving the need to carry out physical disk reads. Flashcache usage (Keep objects) The amount of space (number of megabytes) used for 'keep' objects on the Flashcache. Flashcache usage (total) The number of megabytes used on the Flashcache. Managing Oracle Database Systems: User and Reference Guide 253 Chapter 3—Reference Metric Description MB of Flashcache misses The total amount of data (number of megabytes) read from disks because not all requested data was found in the Flashcache. MB Pushed out of the Flashcache for Keep objects The amount (megabytes) of data pushed out of the Flashcache because the space limit for storing 'keep' objects has been exceeded. MB Pushed out of the Flashcache for Keep objects rate The rate (megabytes/second) at which data was pushed out of the Flashcache because the space limit for storing 'keep' objects has been exceeded. MB read from Flashcache The total amount of data (number of megabytes) from the Flashcache during the specified time range. MB written to Flashcache The total amount of data (number of megabytes) written to the Flashcache during the specified time range. Exadata Cell General Purpose The Exadata Cell General collection retrieves general information and status about the Exadata cell. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Offline 900 Metric Descriptions Metric Description BMC Type Name and version of the Baseboard Management Controller (BMC). Cell ID The Cell ID, provided by the hardware vendor. Managing Oracle Database Systems: User and Reference Guide 254 Chapter 3—Reference Metric Description Cell Model The brand and model of the cell hardware, as provided by the vendor. Cell Name The Exadata cell name. Cell Offload Efficiency The cell's offload efficiency. Cell Offload Processing is a procedure used by the oracle database server to push query predicates (that limit data retrieved by the query using filters and join conditions) to the Exadata cell. Cell Status The Exadata cell status. CPUs The total number of CPU units on the cell. Interconnects The total number of network interconnection cards (NICs). Kernel Version The host kernel's software version. OS Version The Exadata cell's OS version. Power Supplies (Working/Total) The total number of power supplies on the Exadata cell, displayed as working/total. Temperature The server's temperature (Celsius degrees), as provided by the Baseboard Management Controller (BMC). Temperature Status The server's temperature status, as provided by the Baseboard Management Controller (BMC). Total Fans The total number of fans on the Exadata cell. Up Since Calculates the last cell restart or start time, based on the cell_uptime Uptime The time period that has elapsed since the cell was last restarted. Working Fans The total number of working fans on the Exadata cell. Exadata Cell InfiniBand Purpose The Exadata Cell InfiniBand Nodes collection retrieves information and status of the cell InfiniBand nodes. Collection Type Oracle Managing Oracle Database Systems: User and Reference Guide 255 Chapter 3—Reference Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Offline 900 Metric Descriptions Metric Description HCA The InfiniBand node's host channel adapter (HCA) Node IP Address The InfiniBand node's IP address. Node Name The name of the InfiniBand node. Node Status The InfiniBand node's status. Exadata Cell InfiniBand Ports Purpose The Exadata Cell InfiniBand Ports collection retrieves the cell InfiniBand ports’ information and status. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Offline 900 Metric Descriptions Metric Description Link Speed Active The actual link speed of the Exadata cell InfiniBand port, as retrieved in the last sample. Managing Oracle Database Systems: User and Reference Guide 256 Chapter 3—Reference Metric Description Link Speed Enabled The link speed enabled by the user, as retrieved in the last sample. Link Speed Supported The supported link speed of the Exadata cell InfiniBand port, as retrieved in the last sample. Link State The state of the InfiniBand port link of the Exadata cell. Link Width Active The actual link width of the Exadata cell Infiniband port, as retrieved in the last sample. Link Width Enabled The link width enabled by the user, as retrieved in the last sample. Link Width Supported The supported link width of the Exadata cell InfiniBand port, as retrieved in the last sample. Node Name The name of the InfiniBand node. Physical Link State The physical state of the Exadata cell InfiniBand port's link. Port Number The InfiniBand port number. Switch LID The InfiniBand Switch's local identifier (LID), as assigned by the Subnet Manager. Exadata Cell InfiniBand Switches Purpose The Exadata Cell InfiniBand Switches collection retrieves the cell InfiniBand ports’ information and status. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Offline 900 Managing Oracle Database Systems: User and Reference Guide 257 Chapter 3—Reference Metric Descriptions Metric Description Health Details Retrieves the Infiniband health check detailed results of the parameters checked in the environment test. Health Status Retrieves the Infiniband Switch’s current health status, as indicated by the success or failure of the environment test. Link Status The Infiniband Switch’s current link status, as indicated during the last sample. Subnet Manager State The Subnet Manager’s current state on the Infiniband switch. Switch LID The InfiniBand Switch's local identifier (LID), as assigned by the Subnet Manager. Switch Name The Infiniband switch's name. Exadata Cell Statistics Purpose The Exadata Cell Statistics collection retrieves load, throughput and utilization statistics about the Exadata cell. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Offline 900 Metric Descriptions Metric Description Cell ID The Cell ID, provided by the hardware vendor. CPU Busy Percent The percentage of time within the previous minute during which the system CPU units were not idle. Managing Oracle Database Systems: User and Reference Guide 258 Chapter 3—Reference Metric Description KB Received by Ethernet Interfaces Rate The rate (kilobytes/second) at which data was received by the Ethernet interfaces during the last sample. KB Transmitted by Ethernet Interfaces Rate The rate (kilobytes/second) at which data was transmitted by the Ethernet interfaces during the last sample. MB Received by InfiniBand Interfaces Rate The rate (megabytes/second) at which data was received by the InfiniBand interfaces during the last sample. MB Transmitted by InfiniBand Interfaces Rate The rate (megabytes/second) at which data was transmitted by the InfiniBand interfaces during the last sample. Physical Memory Utilization The average percentage of physical memory used on the cell during the last sample. Run Queue Length The average number (over the preceding minute) of processes in the Linux run queue that were marked as running or uninterruptable. This reading is taken from /proc/loadavg. Swap Utilization The percentage of swap used. Exadata Cell General Purpose The Exadata Cell General collection retrieves summarized cell disks performance and throughput statistics. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Offline 900 Managing Oracle Database Systems: User and Reference Guide 259 Chapter 3—Reference Metric Descriptions Metric Description Average Small Read Latency Per Request Average latency in milliseconds to read small blocks per request on all the celldisks. Average Small Write Latency Per Request Average latency in milliseconds to write small blocks per request on all the celldisks. Cell ID The Cell ID, provided by the hardware vendor. MB Read Rate The rate (megabytes/second) at which data was read from all of the celldisks during the specified time range. MB Written Rate The rate (megabytes/second) at which data was written on all of the celldisks during the specified time range. Read Requests Rate Number of requests to read large and small blocks per second on all the celldisks. Write Requests Rate Number of requests to write large and small blocks per second on all the celldisks. Filesystem Disk Purpose The Filesystem Disk collection provides storage definitions regarding the filesystem disk where datafiles, dump files or archive files are located. Collection Type Operating System Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Filesystem size The size of the filesystem Managing Oracle Database Systems: User and Reference Guide 260 Chapter 3—Reference Metric Description Filesystem free % The percentage of the free disk space on the filesystem. Filesystem Free Space The amount (in megabytes) of free disk space on the filesystem Filesystem Name The name of the filesystem Filesystem Type The type of filesystem (datafiles, dump files or mixed files) Long Term Estimated Till Full Estimated number of hours untill the filesystem is full, based on the long-term growth rate Long Term Growth Rate The estimated long-term growth rate of the filesystem Mounted On Mount point name. Operating System The name of the operating system where this instance runs Short Term Estimated Till Full Estimated number of hours until the filesystem is full, based on the short-term growth rate Short Term Growth Rate The filesystem’s estimated short-term growth rate Used The amount of used disk space on the filesystem Used % The percentage of the used disk space on the filesystem Used Growth The growth (in megabytes) of used disk space on the filesystem Flashback Area Purpose If the flash recovery area feature was used for creating a location on disk that enables the database to create various backup and recovery-related files, the Flashback Area collection provides information about this location. Collection Type Oracle Managing Oracle Database Systems: User and Reference Guide 261 Chapter 3—Reference Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 300 Offline 900 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description File Type Indicates the types of the files stored on the flashback area: • ONLINELOG • FLASHBACKLOG • CONTROLFILE • ARCHIVELOG • IMAGECOPY • BACKUPPIECE Non Reclaimable Space Percent The percentage, within the flashback area, of non-reclaimable space (space currently in use, which cannot be reclaimed). Number of Files Indicates the number of files currently stored on the flashback area. Used Space Percent The percentage of used space within the entire disk space allocated to the flashback area. History Locks Tree Instance Purpose The History Locks Tree Instance collection collects snapshots of instance-level locks that reached or exceeded the minimal duration, in seconds, required for collecting the lock. All locks that meet this criterion are displayed in the lower section of the Locks Historical Tree pane. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Frequency Mode Collection Interval (Seconds) Online 60 Offline 60 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Blocked SQL The SQL statement that waits for lock release Blocked SQL Hash The hash value of the statement waiting for lock Blocking Serial Serial number of the locking session Blocking SID A SID that holds locks Client Machine The name of the computer where the client is running Cross Instance Indicates whether the global lock and the locking and blocked sessions are running on different RAC nodes (True) or not (False) DB User Name The Oracle database account of the started session Host The name of the host where the instance is installed Instance Instance name of the session in the lock tree Lock Duration The number of seconds the lock is running Lock Event The wait event on which the blocked statement is waiting Lock Mode The lock mode in which the session holds the lock. The lock mode can have one of the following values: • 0 - none • 1 - null (NULL) • 2 - row-S (SS) • 3 - row-X (SX) • 4 - share (S) • 5 - S/Row-X (SSX) • 6 - exclusive (X) Lock Request Mode The lock mode in which the process requests the lock. The lock request mode has the same possible values as the lock mode. 262 Managing Oracle Database Systems: User and Reference Guide 263 Chapter 3—Reference Metric Description Lock Type The type of user or system lock. The locks on the user types are obtained by user applications. Any process that is blocking others is likely to be holding one of these locks. The user type locks are as follows: • TM - DML enqueue • TX - Transaction enqueue • UL - User supplied Object Wanted The object on which the statement is blocked OS User Name The operating system user for the Client PDB In multitenant architectures, the name of the pluggable database. PID The operating system process ID for the Oracle server process Program The server program name Serial The session’s serial number. The combination of SID and serial number provides a unique identification of the session. SID The session identification number History Locks Tree RAC Purpose The History Locks Tree RAC collection collects snapshot of all locks within the RAC that reached or exceeded the minimal duration, in seconds, required for collecting the lock. All locks that meet this criterion are displayed in the lower section of the Locks Historical Tree pane Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 60 Offline 60 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Blocked SQL The SQL statement waiting for lock release Blocked SQL Hash The hash value of the statement waiting for lock Blocking Serial Serial number of the locking session Blocking SID A SID that holds locks Client Machine The name of the computer where the client is running Cross Instance Indicates whether the global lock and the locking and blocked sessions are running on different RAC nodes (True) or not (False) DB User Name The Oracle database account of the started session Host The name of the host where the instance is installed Instance Instance name of the session in the lock tree Lock Duration The number of seconds the lock is running Lock Event The wait event on which the blocked statement is waiting Lock Mode The lock mode in which the session holds the lock. The lock mode can have one of the following values: • 0 - none • 1 - null (NULL) • 2 - row-S (SS) • 3 - row-X (SX) • 4 - share (S) • 5 - S/Row-X (SSX) • 6 - exclusive (X) Lock Request Mode The lock mode in which the process requests the lock. The lock request mode has the same possible values as the lock mode. Lock Type The type of user or system lock. The locks on the user types are obtained by user applications. Any process that is blocking others is likely to be holding one of these locks. The user type locks are as follows: • TM - DML enqueue • TX - Transaction enqueue • UL - User supplied 264 Managing Oracle Database Systems: User and Reference Guide 265 Chapter 3—Reference Metric Description Object Wanted The object on which the statement is blocked OS User Name The operating system user for the Client PDB In multitenant architectures, the name of the pluggable database. PID The operating system process ID for the Oracle server process Program The server program name Serial The session’s serial number, which, in combination with SID, provides a unique identification of the session. SID The session identification number Host Properties Purpose The Host Properties collection provides operating system version, CPU, memory and network host properties. Collection Type Operating System Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 900 Offline 86400 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Machine Model The model of the host machine (for example: Sun Fire V240) Number of CPU The number of CPU units installed on the host Number of IO Devices The number of IO devices configured on the host Number of Network Interfaces Number of network interfaces installed on the host Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Operating System The name of the operating system installed on the host Operating System Version The version of the operating system installed on the host Physical Hostname The host name, defined per network interface RAM The amount of RAM available on the physical host Initialization Parameters Purpose The Initialization Parameters collection collects definitions and change tracking history of Oracle’s initialization parameters. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 900 Online 900 Offline 3600 Metric Descriptions Metric Description Current Value Parameter value for the session (if the value was modified within the session); otherwise, the instance-wide parameter value. Description The parameter description Instance Level Indicates whether the parameter can have a different value for each instance (TRUE) or must retain the same value for all RAC nodes (FALSE). Is Default? Indicates whether the parameter’s value is set to the default value (TRUE) or was modified (FALSE) 266 Managing Oracle Database Systems: User and Reference Guide 267 Chapter 3—Reference Metric Description Is Deprecated? Indicates whether the parameter has been deprecated (TRUE) or not (FALSE) Is Dynamic? Indicates whether the parameter can be changed with ALTER SESSION (TRUE) or not (FALSE) Is Modified? Indicates whether the parameter has been modified after instance startup: • MODIFIED — the parameter has been modified with ALTER SESSION • SYSTEM_MOD — the parameter has been modified with ALTER SYSTEM, thereby modifying all of the currently logged-in session values • FALSE — the parameter has not been modified after instance startup Is System Modifiable? Indicates whether the parameter can be changed with ALTER SYSTEM and when the change takes effect. This metric can have one of the following values: • IMMEDIATE — the parameter can be changed with ALTER SYSTEM regardless of the type of parameter file used for starting the instance. The change takes effect immediately. • DEFERRED — the parameter can be changed with ALTER SYSTEM regardless of the type of parameter file used for starting the instance. The change takes effect in subsequent sessions. • FALSE — the parameter cannot be changed with ALTER SYSTEM unless a server parameter file was used for starting the instance. The change takes effect in subsequent instances. Parameter Initialization parameter name PDB In multitenant architectures, the name of the pluggable database. Show Hidden Parameters Indicates whether the parameter is hidden (TRUE) or not (FALSE) Type Indicates the parameter type. The possible values are as follows: • Boolean • String • Integer • Parameter file • Reserved • Big integer Managing Oracle Database Systems: User and Reference Guide 268 Chapter 3—Reference Instance General Purpose The Instance General collection collects general instance data such as version and archive logging. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 900 Offline 3600 Metric Descriptions Metric Description DB Version The database version. Is Archived Indicates whether the instance is configured in archive mode. Is Connected to ASM Indicates whether the instance is an Automatic Storage Management (ASM) instance. Is Data Guard Configured Indicates whether DataGuard is configured on this instance. Is Exadata Indicates if this instance is part of Exadata Architecture. Is Flashback Configured Indicates whether the flashback recovery is configured. Is RMAN Configured Indicates whether RMAN is configured on the database for backup and recovery. Invalid Objects Purpose The Invalid Objects collection collects data about invalid objects in each schema. Collection Type Oracle Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Invalid Functions Number of invalid functions in the schema Invalid Package Bodies Number of invalid package bodies in the schema. Invalid Packages Number of invalid packages in the schema. Invalid Procedures Number of invalid procedures in the schema. Invalid Types Number of invalid types in the schema. Invalid Views Number of invalid views in the schema. New Invalid Functions The number of invalid functions that are found to be present in the schema, and were not detected during the last run. New Invalid Package Bodies Number of invalid package bodies that were added to the schema during the last sampling period. New Invalid Packages Number of invalid packages that were added to the schema during the last sampling period. New Invalid Procedures Number of invalid procedures that were added to the schema during the last sampling period. New Invalid Types Number of invalid types that were added to the schema during the last sampling period. New Invalid Views Number of invalid views that were added to the schema during the last sampling period. Others Number of invalid objects of other object types in the schema. PDB In multitenant architectures, the name of the pluggable database. 269 Managing Oracle Database Systems: User and Reference Guide 270 Chapter 3—Reference Metric Description Schema Name The name of the schema that contains the invalid objects Total Invalid Objects Total number of invalid objects in the schema Invalid Objects List Purpose The Invalid Objects List collection lists the invalid Object of the specific type currently detected in the database. Collection Type Oracle Collection Sampling Settings None (on-demand collection). Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Object Name The name of the invalid object Object Type The type of the invalid object PDB In multitenant architectures, the name of the pluggable database. Schema Name The name of the schema that contains the invalid objects IO Activity Purpose The IO Activity collection provides metrics, wait events, and performance ratios related to input/ output activity. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Managing Oracle Database Systems: User and Reference Guide 271 Chapter 3—Reference Frequency Mode Collection Interval (Seconds) Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Avg Read Time The average time (ms) spent reading a data block from the disk into the database buffer cache. Avg Write Time The average time spent (ms) writing a data block from the database buffer to the disk. Balance Average Read Time The extent, in percents, to which average read time is balanced across RAC nodes Balance Average Write Time The extent, in percents, to which average write time is balanced across RAC nodes Balance Cache Hit Ratio The extent, in percents, to which cache hit ratio is balanced across RAC nodes Balance Logical Reads The extent, in percents, to which logical reads are balanced across RAC nodes Balance Physical Reads The extent, in percents, to which physical reads are balanced across RAC nodes Balance Physical Writes The extent, in percents, to which physical writes are balanced across RAC nodes Block Changes per Call This ratio measures the amount of data manipulation within existing database tables, which each transaction performs via DML commands, by computing the average number of blocks changed per single user call. Bea cu se creating or dropping indexes modifies index blocks, such operations increment the value of this metric. Block Reads This rate determines the block get rate, which is a basic measure of the rate at which the application system references the database. Change in metric size may indicate that the database workload has materially increased or that the referenced segment sizes increased. Managing Oracle Database Systems: User and Reference Guide 272 Chapter 3—Reference Metric Description Cache Hit Ratio The effectiveness of the buffer cache, computed by comparing the number of block requests, which found the block in the buffer cache, with the entire block requests (including requests that required disk reads). Chained Row Ratio This ratio measures the percentage of chained or migrated rows encountered during a fetch as a percentage of the entire fetched rows. Consistent Changes The number of times a user process has applied rollback entries to perform a consistent read on the block. Consistent Gets The number of data block reads in CONSISTENT mode (that is, consistent reads). A CONSISTENT read retrieves the block version as of the time the query started. Consistent Gets Direct The number of times a consistent read was carried out directly for a block, bypassing the buffer cache. Consistent Gets From Cache The number of times a consistent read was carried out for a block from buffer cache. DB Block Changes The number of database blocks in memory (SGA) that were modified by the session as a result of DML activity. Such changes generate redo log entries, thereby becoming permanent changes to the database if the transaction is committed. DB Block Gets The number of data block reads in CURRENT mode (that is, current reads). A CURRENT read retrieves the latest version of the block as of the time of reading. Current mode reads are usually associated with DML activity. DB Block Gets Direct Number of times a CURRENT block was retrieved directly, bypassing the buffer cache. DB Block Gets From Cache Number of times a CURRENT block was retrieved from the buffer cache. Disk Sorts Ratio The number of sort operations that require creation of temporary segments on disk to store the intermediate sort results. This occurs if the data being sorted cannot be fit into memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. Managing Oracle Database Systems: User and Reference Guide 273 Chapter 3—Reference Metric Description Full Scan Ratio This ratio measures the percentage of the rows, which were retrieved as a result of full table scans, within the total number of retrieved rows. Oracle Data I/O Operations Total number of I/O operations (both physical and logical) performed by the Oracle session. Oracle Physical Data Processed the total size (in megabytes) of data processed via physical read and write operations during the specified time range. Oracle Physical I/O Operations Total number of physical I/O operations performed by the Oracle session. Physical Read Bytes The number of data blocks read by Oracle, which are not present in the buffer cache or in the PGA, thereby requiring I/ O read operations, that is: the total number of data blocks read from disk. Physical Read IO Requests Number of read requests for application activity (mainly buffer cache and direct load operation) that read one or more database blocks per request. Physical Read Total Bytes Total size in bytes of disk reads by all database instance activity, including application reads, backup and recovery, and other utilities. The difference between this value and the value of the metric physical read bytes provides the total read size in bytes by non-application workload. Physical Read Total IO Requests Number of read requests that read one or more database blocks for all instance activity including application, backup and recovery, and other utilities. The difference between this value and the value of the metric physical read total multi block requests provides the total number of single block read requests. Physical Read Total Multi Block Requests Total number of read requests which read in two or more database blocks per request for all instance activity including application, backup and recovery, and other utilities. Physical Reads The number of data blocks read by Oracle, which are not present in the buffer cache or in the PGA, thereby requiring I/ O read operations, that is: the total number of data blocks read from disk. Physical Reads Cache Total number of data blocks read from disk into the buffer cache. Managing Oracle Database Systems: User and Reference Guide 274 Chapter 3—Reference Metric Description Physical Reads Direct Number of reads that were carried out directly from disk into the PGA, bypassing the buffer cache. Physical Reads Refetch Warmup Number of data blocks that were read from the disk during the automatic pre-warming of the buffer cache. Physical Write Bytes Total size, in bytes, of all disk writes from the database application activity (this metric does not include any other kind of instance activity). Physical Write IO Requests Number of write requests for application activity (mainly buffer cache and direct load operation) that wrote one or more database blocks per request. Physical Write Total Bytes Total size, in bytes, of all disk writes for the database instance, including application activity, backup and recovery, and other utilities. The difference between this value and the value of the metric physical write bytes provides the total write size, in bytes, resulting from non-application workload. Physical Write Total IO Requests Number of write requests which wrote one or more database blocks from all instance activity including application activity, backup and recovery, and other utilities. The difference between this metric value and the value of the metric physical write total multi block requests provides the number of single block write requests. Physical Write Total Multi Block Requests Total number of Oracle instance write requests that wrote two or more blocks per request to the disk for all instance activity, including application activity, recovery and backup, and other utilities. Physical Writes Total number of data blocks written to disk. This metric represents the rate at which Oracle generates system I/O calls. Physical Writes Direct Number of writes directly to disk, bypassing the buffer cache Physical Writes From Cache Total number of data blocks written to disk from the buffer cache Physical Writes Non Checkpoint Number of times a buffer is written for reasons other than checkpoint. This statistic measures the number of writes that would have occurred had no checkpoint operation taken place. Subtracting this value from the value of the metric physical writes provides the number of I/O operations carried out as a result of checkpoint operations. Managing Oracle Database Systems: User and Reference Guide 275 Chapter 3—Reference Metric Description Read Consistency Overhead This ratio measures the extent to which applications must make use of the read consistency mechanism. The queries that processing parts of the UPDATE and/or DELETE operations are subject to read consistency. Read Time The total time spent on reading data from the data file. Reads Per Second Number of physical reads carried out per second Redo Blocks Written Total number of redo blocks written. Dividing this metric by the value of the metric redo writes provides the average number of blocks per write. Redo Buffer Allocation Retries Total number of retries necessary to allocate space in the redo buffer. Retries are needed either because the redo writer has fallen behind or because an event such as a log switch is occurring. Redo Entries The number of redo records created (copied into the redo log buffer). Redo Log Space Requests Number of times the active log file is full and the session must wait for a log switch to be performed. After the completion of the log switch, the redo log entries can be allocated, and the session can be resumed. Redo Log Space Wait Ratio Redo Size Total size, in bytes, of the redo volume. Redo Writes Total number of write operations issued by LGWR to the redo log files. Session Logical Reads The number of data blocks read by Oracle, which are already present in the buffer cache, thereby requiring no I/O read operations. The value of this metric is a summary of the db block gets and consistent gets metrics. Sorts (Disk) Sorts (disk) are sorts that require creation of temporary segments on disk to store the intermediate sort results. Such sorts are carried out if the data being sorted cannot be fit into memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. Managing Oracle Database Systems: User and Reference Guide 276 Chapter 3—Reference Metric Description Sorts (Memory) Number of sort operations that were carried out entirely in memory, without requiring any disk writes. The sort operation can take place in memory if the sorted data can be fit into the memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. Sorts (Rows) The total number of rows sorted. Table Scan Blocks Gotten The number of blocks encountered during the full table scans Table Scan Short Tables The number of times Oracle carried out a complete scan of a short table Table Scans Cache Partitions Number of range scans performed on tables that have the CACHE option enabled Table Scans Direct Read Number of table scans performed using direct read (bypassing the buffer cache) Table Scans Fetch by ROWID Number of rows that were fetched using a ROWID (usually recovered from an index) Table Scans Long Tables The number of times Oracle carried out a complete scan of a long table Table Scans ROWID Ranges The number of table scans conducted during parallel query by the query slaves, with the specified ROWID ranges Table Scans Rows Gotten The number of rows scanned during full table scans. High values may indicate use of non-optimal queries, absence of proper indexing, or both. Tables Fetch Continued Row Number of times a chained or migrated row is encountered during a fetch. Total Physical Data I/O Operations The total number of read and write operations carried out for the data file. Total Sorts Total number of sort operations, including both in-memory and disk sorts. User Calls Number of user calls such as login, parse, fetch, or execute. The value of this metric represents the rate at which the user application addresses Oracle. Managing Oracle Database Systems: User and Reference Guide 277 Chapter 3—Reference Metric Description User Commits Number of transactions committed during the current interval. After a transaction is committed, the redo generated by that transaction reflects the changes made to database blocks, and the database blocks must be written to disk. User Rollbacks Number of transactions that were rolled back, either because of an error or as a result of a manually executed ROLLBACK command Write Time The total time spent on writing data to the datafile Writes Per Second The number of times DBWR was required to write per second. Jobs Queue Purpose The Jobs Queue collection provides jobs queue information and status. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Active Jobs The number of currently running jobs Jobs Idle The number of scheduled jobs that are currently not running Jobs Waiting The number of jobs that are scheduled to be running now but waiting in queue for the resources Total Jobs The total number of scheduled jobs Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Latches Activity Purpose The Latches Activity collection collects latches activity-related data. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Cache Buffer Chain Latch This latch is acquired whenever a block in the buffer cache is accessed (pinned) Cache Buffer LRU Chain Latch This latch is acquired in order to introduce a new block into the buffer cache and when writing a buffer back to disk, specifically when trying to scan the LRU chain that contains all of the dirty blocks in the buffer cache Latch Free Waits Number of waits for a latch that is currently busy (held by another process) Library Cache Latch This latch is acquired in order to add a new statement to the library cache. During a parse, Oracle searches the library cache for a matching statement. If one is not found, then Oracle will parse the SQL statement, obtain the library cache latch and insert the new SQL. Library Cache Pin Latch This latch is acquired when a statement in the library cache is re-executed Redo Allocation Latch This latch is acquired in order to allocate space within the redo log buffer. There is one redo allocation latch per instance. 278 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Redo Copy Latch This latch is acquired in order to write redo records into the redo log buffer. There may be up to LOG_SIMULTANEOUS_COPIES allocation latches. Shared Pool Latch This Latch is acquired when allocating and freeing memory in the shared pool Total Active Time Total amount of time waited for the event (in seconds) Locks Activity Purpose The Locks Activity collection provides locks activity-related data. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Average Lock Duration The average time (in milliseconds) that a resource was locked (held in contention). High values of lock duration may indicate aggressive lock incidence which may be improved through tuning. The metric is calculated as follows: 100 * (Lock Wait) Enqueue Waits Enqueue Waits The total number of times an enqueue request or conversion resulted in a wait. 279 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Enqueue Waits Ratio The percentage of enqueue wait events (waits resulting from enqueue requests or conversions) within the total time consumed by enqueue operations. Gets The total number of enqueue requests or enqueue conversions for this type of enqueue. Gets Pct The percentage of lock requests of this type within the total number of lock requests Lock Efficiency The percentage of lock requests that did not escalate to a blocking lock. Lock Name The name of the lock. Lock Wait The total amount of time spent waiting for the enqueue or enqueue conversion. Waits Pct The percentage of waits for this type of lock, of all the locks. Locks Breakdown Purpose The Locks Breakdown collection provides lock requests and wait information, including the percentage of time spent on locks of the following types: system, TX, TM, User-defined. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 60 Offline 300 280 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description PDB In multitenant architectures, the name of the pluggable database. System Locks Percentage The percentage of time the sessions spent waiting on System locks, out of the entire time spent waiting on locks. System Locks Wait Time The percentage of time the sessions spent waiting on System locks. TM Locks Wait Time The amount of time the sessions spent waiting on TM locks. TM Percentage The percentage of time the sessions spent waiting on TM locks, out of the entire time spent waiting on locks. Total Lock Wait The total time the sessions spent waiting on locks. TX Locks Wait Time The amount of time the sessions spent waiting on TX locks. TX Percentage The percentage of time the sessions spent waiting on TX locks, out of the entire time spent waiting on locks. UL Locks Wait Time The amount of time the sessions spent waiting on user-defined lock types. UL Percentage The percentage of time the sessions spent waiting on userdefined locks, out of the entire time spent waiting on locks. Locks Instance Purpose The Locks Instance collection provides summarized lock requests and application efficiency information. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 60 Offline 300 281 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Balance Enqueue Waits The extent, in percents, to which enqueue wait events are balanced across RAC nodes Balance Lock Efficiency The extent, in percents, to which lock efficiency is balanced across RAC nodes Balance Lock Wait The extent, in percents, to which lock wait events are balanced across RAC nodes Enqueue Waits Total number of waits on enqueues Lock Efficiency The percentage of total locks waits within the total lock requests Locks Tree Purpose The Locks Tree Instance collection provides lock tree information at the instance level. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Blocked SQL The SQL statement waiting for lock release. Blocking Serial The session serial number is either: • Waiting for lock (blocked). • Holding the object (locking). Blocking SID The session identifier for the session is holding the object (locking). Client Machine The name of the Client machine Cross Instance The lock is between different cluster nodes for global locks. 282 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description DB User Name The Oracle user, which is either: • Waiting for lock (blocked) • Holding the object (locking) Host The host where the Oracle instance is installed, which is either: • Waiting for lock (blocked) • Holding the object (locking) Instance The Oracle instance session is either: • Waiting for lock (blocked). • Holding the object (locking). Lock Duration The wait length of time, in seconds. Lock Event The event on which the blocked session is waiting Lock Mode The lock mode in which the session holds the lock Lock Request Mode The lock mode in which the process requests the lock Lock Type The type of user or system lock. Object Wanted The object requested for lock. OS User Name The OS user name with which the session logged in is either: • Waiting for lock (blocked). • Holding the object (locking). PDB In multitenant architectures, the name of the pluggable database. PID The operating system process ID for the Oracle server process. The process ID is either: • Waiting for lock (blocked) • Holding the object (locking) Program The operating system program is either: • Waiting for lock (blocked) • Holding the object (locking) Serial The session serial number is either: • Waiting for lock (blocked). • Holding the object (locking). SID The blocked session’s system ID (SID) 283 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Locks Tree RAC Purpose The Locks Tree RAC collection provides lock tree information at the global level. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 seconds Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Blocked SQL The SQL statement waiting for lock release Blocking Serial The session serial number is either: • Waiting for lock (blocked). • Holding the object (locking). Blocking SID The session identifier for the session is holding the object (locking). Client Machine The name of the host where the instance is running Cross Instance The lock is between different cluster nodes for global locks. DB User Name The Oracle user (locking/blocked) that is either: • Waiting for lock (blocked). • Holding the object (locking). Host The host where that the Oracle instance is installed is either: • Waiting for lock (blocked). • Holding the object (locking). Instance The instance name of the session in the lock tree Lock Duration The wait length of time, in seconds. Lock Event The wait event on which the blocked session is waiting Lock Mode The lock mode in which the session holds the lock. Lock Request Mode The lock mode in which the process requests the lock. 284 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Lock Type The type of user or system lock. Object Wanted The object requested for lock. OS User Name The OS user name that the session logged in with is either: • Waiting for lock (blocked). • Holding the object (locking). PDB In multitenant architectures, the name of the pluggable database. PID The operating system process ID for the Oracle server process. The process ID is either: • Waiting for lock (blocked). • Holding the object (locking). Program The operating system program name is either: • Waiting for lock (blocked). • Holding the object (locking) SID The blocked session’s system ID (SID) OS and Instance Statistics Purpose The OS and Instance Statistics collection provides the following data: • CPU and memory information about the operating system • CPU and memory-related information about the operating system and the monitored instance. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 seconds Online 60 seconds Offline 300 seconds 285 Managing Oracle Database Systems: User and Reference Guide 286 Chapter 3—Reference Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description ASM Buffer Cache Size The amount of memory allocated to the ASM buffer cache. Automatic SGA Indicates whether the database is configured in manual or automatic memory management mode. Average Connected Sessions Total number of current logons. Balance System CPU Utilization The extent (displayed as a percentage) to which system CPU utilization is balanced across RAC nodes. Buffer Cache Size The amount of memory currently allocated to the buffer cache. The buffer cache, which is used for caching frequently accessed data blocks, is normally the largest memory cache used by Oracle. Bytes received via SQL*Net from dblink Total number of bytes received from a database link over Oracle Net Services. Bytes sent via SQL*Net to client Total number of bytes sent to the client from the foreground processes. Bytes sent via SQL*Net to dblink Total number of bytes sent over a database link. Calls Measures the workload being placed on the instance from all work sources. Current SGA Size The actual size currently allocated to the System Global Area (SGA) memory. Cursor Authentications Number of privilege checks conducted during execution of an operation. Data Dictionary Cache Hit Ratio The data dictionary cache hit ratio is used for measuring the ratio of dictionary hits to misses. The data dictionary cache is used for holding rows from the internal Oracle metadata tables, including SQL stored in packages. Default 16K Buffer Cache Size The default space allocated within the buffer cache to 16k-size data blocks. Default 2K Buffer Cache Size The default space allocated within the buffer cache to 2k-size data blocks. Managing Oracle Database Systems: User and Reference Guide 287 Chapter 3—Reference Metric Description Default 32K Buffer Cache Size The default space allocated within the buffer cache to 32k-size data blocks. Default 4K Buffer Cache Size The default space allocated within the buffer cache to 4k-size data blocks. Default 8K Buffer Cache Size The default space allocated within the buffer cache to 8k-size data blocks. Default Buffer Cache Size Size in MB of default cache area of SGA that cache blocks of data and indexes that are not assigned to specific buffer pools. Execute Count Number of statements executed during the specified time range. Fixed Size The fixed part of the SGA is an internal area set by Oracle, which cannot be manually modified. This area contains the following data: • General information about the state of the database and the instance that needs to be accessed by background processes • Information communicated between processes Flash Cache Insert Skip: DBWR Overloaded The number of writes to the flash cache that were skipped (insert skip) as a result of an overloaded DBWR, which has exceeded a predefined threshold. Flash Cache Read Hit Ratio The percentage of read requests that were resolved from Flash Cache, out of the total of physical read total IO requests. Flash Cache Read Hits The number of read requests, resolved from Flash Cache. Free Memory Amount of free RAM in the system. Free SGA Pct The percentage of free memory out of the total amount of memory allocated to SGA. Free SGA Size The size (in megabytes) of free memory out of the total amount of memory allocated to SGA. Hard Parse Ratio The ratio of hard parse calls (real parses) to total parses. Instance CPU The amount of CPU time used by a session from the time a user call starts until it ends. Instance CPU Load The amount of time the instance was using CPU within the total elapsed time (as a percentage). Managing Oracle Database Systems: User and Reference Guide 288 Chapter 3—Reference Metric Description Instance RAM Percentage Total memory, in percentage, allocated to the instance out of the total RAM. Java Pool Free Pct The amount of free space in the Java pool, expressed as a percentage of the Java pool’s total size. Java Pool Free Size The amount (in megabytes) of free space in the Java pool. Java pool Size The size of the Java pool, which caches class definition, Java methods, and Java objects. An improperly sized Java pool can affect the parse rate. Java Pool Used Size The amount (in megabytes) of used space in the Java pool. Keep Pool Cache Size The size of the Keep buffer pool. This pool contains frequently accessed segments in the buffer cache. Large Pool Free Pct The amount of free space in the Java pool, expressed as a percentage of the large pool’s total size. Large Pool Free Size The amount (in megabytes) of free space in the large pool. Large Pool Size The size of the Large pool. This pool is used in multi-threaded servers (MTS) for session memory, parallel execution, and backup processes. Large Pool Used Percent The percentage of used space in the large pool. Large Pool Used Size The amount (in megabytes) of used space in the large pool. Library Cache Hit Ratio The effectiveness of the library cache, computed by comparing the number of queries submitted to Oracle server for execution, which found the parsed execution plan in the library cache (soft parsing), with the entire number of queries submitted during the specified time range (including queries that required the parsing of a new statement, also known as "hard parsing"). Logons Cumulative The total number of current logons. Logons per Second The average number of logons per second for the specified time range. Maximum Memory Size The maximum size (in megabytes) of memory that can be allocated to an Oracle instance. Maximum SGA Size The maximum defined size of the System Global Area (SGA) memory. Managing Oracle Database Systems: User and Reference Guide 289 Chapter 3—Reference Metric Description Memory Usage The amount of memory consumed by all OS processes (including Oracle). Memory Utilization The relative share, in percents, of memory consumed by all OS processes (including Oracle) out of the total memory. Messages Received Number of messages sent and received between background processes. Opened Cursors Cumulative Total number of cursors opened since the instance started. Opened Cursors Current Total number of currently open cursors. Oracle/System CPU Usage The amount of CPU consumed by the entire system, broken down into monitored instance activity and all other activities Parse count (hard) The number of times an SQL statement has undergone a hard parsing process, which requires reloading the SQL statement to the shared pool, along with the associated RAM memory allocation. Parse count (total) Total number of parse calls (hard and soft). PDB In multitenant architectures, the name of the pluggable database PGA Currently Allocated The amount of Program Global Area (PGA) memory allocated for the current session. Physical Read Total IO Requests The total number of I/O requests due to physical reads. Potential SGA Growth The amount (in megabytes) by which the SGA memory has increased during the specified time range. Recursive Calls Number of recursive calls generated at both the user and system level. For more information, see “Recursive Calls” on page 348. Recycle Pool Cache Size The amount of memory currently allocated to the Recycle buffer pool. This pool contains blocks belonging to segments that are not frequently accessed. Redo Buffers Cache Size The size of the redo buffer, which contains redo entries that must be written to the redo log. A properly sized redo buffer improves performance by minimizing the time waiting for writes to the log file to complete. Managing Oracle Database Systems: User and Reference Guide 290 Chapter 3—Reference Metric Description Result Cache Find/Create Ratio The result cache find/create ratio is used for measuring the ratio of result cache hits to misses. The Result Cache is an area within the shared pool, which contains the end results of a query execution. Failure to find these end results, that is: a result cache miss, leads to creation operation. Low values of this metric indicate that the contents of the result cache are being inefficiently used. Result Cache Max Size The maximum size (in megabytes) of memory allocated to the result cache. Session Cursor Cache Count Total number of cursors cached. This statistic is incremented only if SESSION_CACHED_CURSORS > 0. Session Cursor Cache Hits Number of hits in the session cursor cache. A hit means that the SQL statement did not have to be re-parsed. Shared Pool Free Pct The amount of free space in the shared pool, expressed as a percentage of the shared pool’s total size. Shared Pool Free Size The amount (in megabytes) of free space in the shared pool. Shared Pool Size The size of the shared pool. The shared pool caches SQL statements, PL/SQL programs, object definitions and session memory for MTS sessions. Performance can improve when the shared pool is sized properly, thereby avoiding reparsing and reloading. Shared Pool Used Percent The amount of used space in the shared pool, expressed as a percentage of the shared pool’s total size. Shared Pool Used Size The amount (in megabytes) of used space in the shared pool. SQL*Net roundtrips to/ from client Total number of Net8 messages sent to and received from the client SQL*Net roundtrips to/ from dblink Total number of Net8 messages sent and received over a database link. Stream Pool Size The size (in megabytes) of the stream pool area (the area that caches Oracle Streams objects) within the SGA. System CPU Utilization The overall operating system CPU usage, including Oracle. System Kernel CPU Utilization Percentage of CPU time consumed by the operating system's processes (kernel mode activities). Managing Oracle Database Systems: User and Reference Guide 291 Chapter 3—Reference Metric Description System User CPU Utilization Percentage of CPU time consumed by the operating system's processes (user mode activities). Target Memory Size The size (in megabytes) required for optimal memory utilization by the monitored Oracle instance. When this parameter, which was introduced in Oracle 11g, is set, Oracle automatically tunes memory allocation by resizing the SGA and PGA as needed. Target SGA Size Specifies the total size (in megabytes) of all SGA components. Specifying the SGA_TARGET results in automatic sizing of the following memory pools: • Buffer cache • Shared pool • Large pool • Java pool • Streams pool Transactions The number of transactions running. Used RAM Pct Rate The overall utilization of RAM by all processes, including Oracle. Used SGA Pct The amount of used space in the SGA memory, expressed as a percentage of the SGA memory’s total size. Used SGA Size Total size of the currently used SGA memory. User Calls Number of user calls such as login, parse, fetch, or execute. This metric value represents the rate at which the user application addresses Oracle. User Calls Per Parse The average number of calls in relation to statement parses. Changes in this rate most likely result from application change, but may also indicate that usage patterns are changing and users are moving from one module to another, either more or less frequently. User Commits Number of transactions committed during the current interval. After a transaction is committed, the redo generated by that transaction reflects the changes made to database blocks and they must be written to disk. User Rollback Ratio The percentage of rolled back transactions, that is: transactions for which the ROLLBACK statement was issued, either manually or as a result of an error that occurred during the transaction. Managing Oracle Database Systems: User and Reference Guide 292 Chapter 3—Reference Metric Description User Rollbacks Number of transactions that were rolled back. Rollbacks occur either because of a manually issuing a ROLLBACK command or in the event of an error. Variable Others Size The size of variable portion of SGA excluding shared pool, large pool and java pool. Variable Size The variable portion of SGA, which contains shared pool, large pool and java pool. OS File System IO Summary Purpose The OS File System IO Summary collection provides IO information about the operating system. Collection Type Operating System Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 seconds Online 60 seconds Offline 300 seconds Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Operating System The name of the operating system where the selected instance is running. System Physical Reads The number of physical read operations carried out from this operating system. System Physical Writes The number of physical write operations carried out from this operating system. System RW Operations Total number of disk operations (both read and write) carried out by the operating system. This metric does not include operations that were satisfied using the file system. Managing Oracle Database Systems: User and Reference Guide 293 Chapter 3—Reference OS General Network Summary Purpose The OS General Network Summary collection provides network information about the operating system. Collection Type Operating System Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Net Collisions The number of times two computers sent packets at the same time on the network, and the packets “collided”, resulting in the need to retransmit both packets. Net Errors Number of errors reported on the various network interfaces. Network Name The network interface name. Operating System The name of the operating system where the selected instance is running. Packets The total sum of packets received and sent on this network interface (the sum of the packets_in and packets_out values). Packets In Number of packets received by this network interface. Packets Out Number of packets sent from this network interface. PA Instance Statistics Purpose The PA Instance Statistics collection collects metrics regarding the monitored instance, which are only available for a cartridge with the Performance Analysis agent installed and configured. Managing Oracle Database Systems: User and Reference Guide 294 Chapter 3—Reference Collection Type Performance Analysis Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Average SQL Active Time Per Call This ratio measures the net average Active Time for a given SQL cursor per Oracle call (includes both user calls and recursive calls). Active time is measured from cursor creation until the cursor is closed. Each cursor execution generates several calls depending on the cursor type and the resulting set size (in the case of SELECT statements). Average SQL Response Time The gross response time for a given SQL cursor sampled during time interval; response time is measured beginning with cursor creation and continuing until the cursor is closed. Each cursor execution generates several calls depending on the cursor type and the resulting set size (in the case of SELECT statements). Degree Of Parallelism Average number of server processes (master and slaves) assigned to handle a single operation. Elapsed Time The amount of time (in microseconds) used by the active cursor for parsing/executing/fetching. Enqueue Requests Total number of table or row locks obtained by the session. Non Oracle CPU Usage Overall CPU consumption not associated with the monitored Oracle instance (background and foreground processes). Note This value includes CPU consumption generated by other Oracle instances running on the monitored instance host. Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Non Oracle Resident Memory Usage Amount of physical RAM consumed by processes other than that of the monitored Oracle instance. Oracle BG CPU Usage CPU time consumed by the monitored Oracle instance’s background processes. Oracle BG Resident Memory Usage Amount of physical memory consumed by Oracle background processes. Oracle CPU Usage CPU time consumed by all of the monitored Oracle instance’s processes, both background and user processes. Oracle FG CPU Usage CPU time consumed by the monitored Oracle instance’s user (session) processes. Oracle FG Memory Usage Amount of physical and swap memory consumed by Oracle foreground (session) processes. Oracle FG Resident Memory Usage Amount of physical memory consumed by the monitored Oracle instance foreground (session) processes. Oracle Resident Memory Usage The amount of physical memory consumed by the various monitored Oracle instance processes (including the SGA). Resident SGA Ratio This ratio measures the relative SGA portion, which is RAM resident. Resident SGA Size Amount of SGA resident in physical memory. In a properly working environment, all SGA is RAM resident. When portions of SGA are being swapped, Oracle performance is significantly degraded, as previously cached data now requires I/O to be accessed. Samples Number of Collector SGA samples performed during the interval. Session Logoffs Number of sessions logged out during the current interval. Session PGA Memory Amount of PGA memory allocated to the session. Session UGA Memory Amount of UGA memory allocated to the session. 295 Managing Oracle Database Systems: User and Reference Guide 296 Chapter 3—Reference Metric Description SGA Swap Size Amount of swapped SGA. In a properly working environment, this metric's value should be zero. When portions of SGA are being swapped, Oracle performance is significantly degraded, as previously cached data now requires I/O to be accessed. SQL Executions Ended Number of statements whose activity finished during the current interval. SQL Executions Started Number of statements issued during the current interval. Total Memory Usage Amount of memory consumed by entire operating system processes (including Oracle), both physical and swap. PA Usability Purpose The PA Usability collection collects data about the state of the Performance Analysis agent and its components. Collection Type Performance Analysis Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 300 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Collector State The state of the Performance Analysis agent’s collector. Collector State Availability The amount of time that the Performance Analysis agent was active during the specified time range. Last Error Message The error message that was last issued by Performance Analysis. Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Last Update Date The date when the most recent update of the Performance Analysis agent took place. Middleware state The state of the Performance Analysis agent’s collector. Up Since The time when the Performance Analysis agent started. Parallel Purpose The Parallel collection provides information about the instance’s parallel executions. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Active Parallel Sessions (11i) Number of active parallel sessions Active Serial Sessions (11i) Number of active sessions that were serialized DDL Statements Parallelized Number of DDL statements that were executed in parallel DML Statements Parallelized Number of DML statements that were executed in parallel Parallel operations Downgraded Total number of times parallel execution was requested but execution was downgraded 297 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Parallel Operations Downgraded 1 to 25 Pct Number of times parallel execution was requested but execution was downgraded up to 25% due to insufficient parallel execution servers Parallel Operations Downgraded 25 to 50 Pct Number of times parallel execution was requested but execution was downgraded 25% or more as a result of insufficient parallel execution servers Parallel Operations Downgraded 50 to 75 Pct Number of times parallel execution was requested but execution was downgraded 50% or more as a result of insufficient parallel execution servers Parallel Operations Downgraded 75 to 99 Pct Number of times parallel execution was requested but execution was downgraded 75% or more as a result of insufficient parallel execution servers Parallel Operations Downgraded to Serial Number of times parallel execution was requested but execution was serial as a result of insufficient parallel execution servers Parallel Operations not Downgraded Number of times parallel execution was executed at the requested degree of parallelism Parallel QC (11i) The number of parallel query coordinators Parallel Slaves (11i) The number of parallel slaves PQ Servers Busy The number of servers that were simultaneously executing SQL queries and DML statements. PQ Servers Busy Percent The percent of busy parallel servers within the total number of parallel servers PQ Servers Idle The number of currently idle parallel servers PQ Servers Limit The maximum allowed number of parallel servers Queries Parallelized Number of SELECT statements executed in parallel PGA Statistics Purpose The PGA Statistics collection collects information about the use of the PGA memory. Collection Type Oracle 298 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description PGA Aggregate Target The value of the PGA Aggregate Target parameter in bytes. The pga_aggregate_target parameter, which was introduced in Oracle 11g, defines the total amount of memory available in Oracle for allocation to the PGA memory structure. If the value of this parameter is set correctly, all work areas executed by the system since start-up are processed exclusively in the PGA memory, requiring no extra passes. In such a case, the PGA hit percent’s value is 100%, and Oracle does not need to allocate extra memory to the PGA. However, when this value is set too small, the hit percent’s value decreases, and Oracle may need to perform one or more extra memory allocations. PGA Hit Percent The PGA cache hit ratio (as a percentage). When the value of this metric is 100%, it indicates that all work areas executed by the system processed the entire input data exclusively via the PGA memory, thereby making optimal use of this memory. Any lower value indicates that at least one work area required one or more extra passes over the input data. 299 Managing Oracle Database Systems: User and Reference Guide 300 Chapter 3—Reference Metric Description PGA Overallocation Count The number of extra PGA memory allocations that were carried out during the specified time range. Over-allocating PGA memory can happen if the value of the dynamic initialization parameter, PGA_AGGREGATE_TARGET, which is set to all server processes, is too small to allow executing all work area workload (operations) exclusively in the PGA memory. Such a scenario requires Oracle to allocate extra PGA memory. Ideally, the PGA cache hit ratio should be close to 100%, resulting in an extremely small number (close to zero) of PGA overallocations. High values of the PGA Overallocation Count metric indicate the need to increase the value of PGA_AGGREGATE_TARGET. Pluggable Databases Information Purpose This collection contains general information about pluggable databases. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 3600 Online 3600 Offline 3600 Metric Descriptions Metric Description PDB ID The identifier used for the pluggable database. PDB The name of the pluggable database. Cloned From PDB Name Indicates from which pluggable database the selected pluggable database is cloned. Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Created Time The time when the pluggable database was created. Share The portion of the system's CPU resources that is allocated to the pluggable database. Pluggable Databases State Purpose This collection contains the status, state, and startup time for pluggable databases. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 900 Offline 900 Metric Descriptions Metric Description PDB ID The identifier used for the pluggable database. PDB The name of the pluggable database. PDB Availability Status of the pluggable database, expressed as a percentage. Open Mode State of the pluggable database. Open Time Startup time of the pluggable database. Redo Archive Performance Purpose The Redo Archive Performance collection provides redo and archive activity information. Collection Type Oracle 301 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Average Redo Write Time The average time (ms) spent writing a redo log entry to the log files. User sessions experience log write waits when, upon commits, an individual session waits for the LGWR to flush its redo information to the redo log file. Balance Archive Waits The extent to which archive wait events are balanced across RAC nodes. Balance Average Redo Write Time The extent to which average redo write time is balanced across RAC nodes. Balance Redo Waits The extent to which average redo wait events are balanced across RAC nodes. Balance Redo Writes The extent to which redo writes are balanced across RAC nodes. Commits Number of transactions committed during the current interval. After a transaction is committed, the redo generated by that transaction reflects the changes made to database blocks, which must be written to disk. Log File Sync Event Average time spent on writing redo log entries during a commit. Log Switch Wait The time the session spent waiting for the completion of a redo log switch. Log Write Wait Time spent waiting for the completion of a write (physical I/O) to a logfile. LGWR experience log write waits when it periodically flushes the contents of the log buffer into the active log file. 302 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Redo Waits Elapsed redo wait time, including the time spent waiting on contention resulting from: • User sessions writing redo entries to the redo log buffer • LGWR writing the log buffer contents into the active redo log file • ARCH archiving redo logs to the destination location Redo Writes Total number of write operations issued by the log writer (LGWR) to the redo log files. Redo Writes Volume Total amount (in megabytes) of redo operations generated and written to redo logs Time to Sync a Redo Log Entry Average time spent on writing single redo log entry to logfile during a commit. Redo Instances Miscellaneous Purpose The Redo Instances Miscellaneous collection provides Redo Log Switches information. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 3600 Online 3600 Offline 3600 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Log Switches The number of redo Logs switches during the specified time range 303 Managing Oracle Database Systems: User and Reference Guide 304 Chapter 3—Reference Redo Logs Member Storage Purpose The Redo Logs Member Storage is a list of Redo log members and their properties. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 900 Offline 3600 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Group Number The redo log group number Member Name The redo log group member name Member Status The redo log group member status is one of the following: • invalid • stale • deleted • in use Redo Log Group Redo Log Group name Redo Log Storage Purpose The Redo Log Storage collection provides a list of all Redo log groups and their overall properties. Collection Type Oracle Managing Oracle Database Systems: User and Reference Guide 305 Chapter 3—Reference Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 900 Offline 3600 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Archived The archive status of the log file: • yes • no First change# A system-allocated number (SCN), identifying the first time a change was recorded in this log. First Time The time at which the first SCN in the log was recorded. Group Status The status of the group is one of the following: • unused • current • clearing • clearing_current • inactive This information is derived from the v$log table. Group# The log group number. Sequence# The sequence number. Size The size (in megabytes) of the log . Session Details Datafile I/O PA Purpose The Session Details Datafile I/O PA collection, which is active only if the Performance Analysis agent is installed and configured, uses this agent to retrieve SQL summary information regarding Current and Top Sessions. Collection Type PA/Oracle Managing Oracle Database Systems: User and Reference Guide 306 Chapter 3—Reference Collection Sampling Settings None (on-demand collection). Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Files The datafile name. I/O Wait Time spent for the completion of input/output operations. Serial Number The session’s serial number. The combination of SID and serial number provides a unique identification of the session. Session ID The session identification number (SID). Session Details Locks PA Purpose The Session Details Locks PA collection retrieves detailed lock information regarding Current and Top Sessions, via the Performance Analysis agent, when available. Collection Type PA/Oracle Collection Sampling Settings None (on-demand collection). Metric Descriptions Metrics that refer to the blocked session \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Serial Number The session’s serial number. The combination of SID and serial number provides a unique identification of the session. Session ID The blocked session's identification number Metrics that refer to the blocking session Metric Description DB User The Oracle database account of the blocking session. Duration from Start The lock's total duration from its start time, which is indicated in the Lock Time field. Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Lock Time The exact time at which the lock started. Object Wanted The object requested for lock. Serial Number The session’s serial number. The combination of SID and serial number provides a unique identification of the session. Serial[j] The blocking session’s serial number. Server PID The blocking server’s server ID Session ID The session identification number. SID The blocking session's identification number. Session Details Open Cursors PA Purpose The Session Details Open Cursors PA collection retrieves open cursor information via the Performance Analysis agent, when available. Collection Type PA/Oracle Collection Sampling Settings None (on-demand collection). Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Active Time The amount of time, in seconds, spent on wait events. Buffer Wait The percentage of time spent on Buffer Wait events within the total active time. Cluster Wait The percentage of time spent on Cluster Wait events within the total time the session spent on wait events. CPU Usage The percentage of CPU Usage time within the total active time. CPU Wait The percentage of time spent on CPU Wait events within the total active time. Hash Value The hash value of the statement waiting for lock 307 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description I/O Wait The percentage of time spent on I/O Wait events within the total active time. Latch Wait The percentage of time spent on Latch Waitevents within the total active time. Lock Wait The percentage of time spent on Lock Wait events within the total active time. Network Wait The percentage of time spent on Network Wait events within the total active time. Other Wait The percentage of time spent on Other Wait events within the total active time. Serial Number The session’s serial number. The combination of SID and serial number provides a unique identification of the session. Session ID The session identification number. SQL Text First n characters of the SQL text for the current cursor. Session Details SQL Summary PA Purpose The Session Details SQL Summary PA collection retrieves datafile I/O information via the Performance Analysis agent, when available. Collection Type PA/Oracle Collection Sampling Settings None (on-demand collection). Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Active Time Time waited for the event (in seconds) Buffer Wait The percentage of time spent on Buffer Wait events within the total active time. Cluster Wait The percentage of time spent on Cluster Wait events within the total time the session spent on wait events. 308 Managing Oracle Database Systems: User and Reference Guide 309 Chapter 3—Reference Metric Description CPU Usage The percentage of CPU Usage time within the total active time. CPU Wait The percentage of time spent on CPU Wait events within the total active time. Hash Value The hash value of the statement waiting for lock. I/O Wait The percentage of time spent on I/O Wait events within the total active time. Latch Wait The percentage of time spent on Latch Wait events within the total active time. Lock Wait The percentage of time spent on Lock Wait events within the total active time. Network Wait The percentage of time spent on Network Wait events within the total active time. Other Wait The percentage of time spent on Other Wait events within the total active time. Serial Number The session’s serial number. The combination of SID and serial number provides a unique identification of the session. Session ID The session identification number. SQL Text First n characters of the SQL text for the current cursor. SQL Executions Number of statements sampled during the specified time range. Session Details Statistics PA Purpose The Session Details Statistics PA collection retrieves detailed statistics regarding Top Sessions via the Performance Analysis agent, when available. Collection Type PA/Oracle Collection Sampling Settings None (on-demand collection). Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Average Lock Duration The average length of time for a lock, in milliseconds. Average Read Time The average time (ms) spent reading a data block from the disk into the database buffer cache. Consistent Changes The number of times a user process has applied rollback entries to perform a consistent read on the block. Consistent Gets The number of data block reads in CONSISTENT mode (that is, consistent reads). A CONSISTENT read retrieves the block version as of the time the query started. Current Reads The number of data block reads in CURRENT mode. A CURRENT read retrieves the latest version of the block as of the time of reading. Current mode reads are usually associated with DML activity. DB Block Changes The number of database blocks in memory (SGA) that were modified by the session as a result of DML activity. Such changes generate redo log entries, thereby becoming permanent changes to the database if the transaction is committed. Degree of Parallelism The total number of server processes (master and slaves) assigned to handle a single operation. Enqueue Requests The total number of table or row locks acquired per transaction. Enqueue Waits The total number of times an enqueue request or conversion resulted in a wait. Kbytes Sent Via SQL*Net To Client Total number of kilobytes sent to the client from the foreground processes Kbytes Sent Via SQL*Net To DBLink Total number of kilobytes sent over a database link Logical Reads The number of logical reads carried out by the session since logon Parse Count The number of parse request performed by this session. 310 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Physical Reads The number of data blocks read by Oracle, which are not present in the buffer cache or in the PGA, thereby requiring I/ O read operations, that is: the total number of data blocks read from disk. Physical Reads Direct Number of reads that were carried out directly from disk into the PGA, bypassing the buffer cache. Physical Writes Direct Number of writes directly to disk, bypassing the buffer cache. Recursive Calls Number of Recursive Calls generated at both the user and system level. Redo Entries The number of redo records created (copied into the redo log buffer). Serial Number The session’s serial number. The combination of SID and serial number provides a unique identification of the session. Session ID The session identification number. Sort Rows The total number of rows sorted. Sorts Disk Sorts that require creation of temporary segments on disk to store the intermediate sort results. Such sorts are carried out if the data being sorted cannot be fit into memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. Sorts In Memory Sort operations that were carried out entirely in memory, without requiring any disk writes. The sort operation can take place in memory if the sorted data can be fit into the memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. SQL*Net Roundtrips To/ From Client Total number of *Net messages sent to and received from the Client. Table Fetch By ROWID Number of rows that were fetched using a ROWID (usually recovered from an index). Table Fetch Continued Row Number of times a chained or migrated row is encountered during a fetch. Table Scans Long Tables The number of times Oracle carried out a complete scan of a long table. 311 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Table Scans ROWID Ranges The number of table scans conducted during parallel query by the query slaves, with the specified ROWID ranges. Table Scans Rows Gotten The number of rows scanned during full table scans. High values may indicate use of non-optimal queries, absence of proper indexing, or both. Table Scans Short Tables The number of times Oracle carried out a complete scan of a short table. User Calls Number of user calls such as login, parse, fetch, or execute. The value of this metric represents the rate at which the user application addresses Oracle. Session List PA Purpose The Session List PA collection retrieves Current and Top Sessions information via the Performance Analysis agent, when available. Collection Type PA/Oracle Collection Sampling Settings None (on-demand collection). Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Action The currently executing action name set by DBMS_APPLICATION_INFO.SET_MODULE Active Time The amount of time, in seconds, spent on wait events. Archival Any activity that generates a LGWR (log writer) for archiving the redo logs to the target location. Additionally, user sessions can also experience archival waits when the login to which the LGWR is going to switch has not yet been archived and the session is issuing DML statements. Average Lock Duration The average length of time for a lock, in milliseconds. Average Read Time The average time (ms) spent reading a data block from the disk into the database buffer cache. 312 Managing Oracle Database Systems: User and Reference Guide 313 Chapter 3—Reference Metric Description Average SQL Active Time Per Call This ratio measures the net average Active Time for a given SQL cursor per Oracle call (includes both user calls and recursive calls). Active time is measured from cursor creation until the cursor is closed. Each cursor execution generates several calls depending on the cursor type and the resulting set size (in the case of SELECT statements). Average SQL Response Time The gross response time for a given SQL cursor sampled during time interval; response time is measured beginning with cursor creation and continuing until the cursor is closed. This value tends to be higher than the value of average Active Time Per Execution, since Active Time measures the net SQL activity; net SQL activity ignores any pauses in SQL execution such as sporadic application processing between distinct Fetch operations. Average Write Time Average time spent on a single physical write request for a specific datafile. Backup/Recovery When archiving is not performed fast enough, database operations can be suspended temporarily whenever the log writer is forced to wait for an inactive group to become available for reuse. BG CPU Time The amount of CPU time used by background processes. Block Changes Per Call This ratio measures the amount of DML work that each transaction performs by computing the average blocks changed per single user call. Creating or dropping indexes affects this value, because changes to index blocks increment it. Block Reads This rate determines the block get rate, which is a basic measure of the rate at which the application system references the database. Change in metric size may indicate that the database workload has materially increased or that the referenced segment sizes increased. Blocking Session Status The status of the blocking session; indicates whether the session is currently executing SQL statements (ACTIVE) or not (INACTIVE). Buffer Busy Wait The time spent by the session when trying to access a data block that is unavailable for one of the following reasons: • The data block is currently not in memory, but another process has already issued an I/O request to read it into memory • The data block is in memory, but in an incompatible mode (for example, current instead of consistent) Managing Oracle Database Systems: User and Reference Guide 314 Chapter 3—Reference Metric Description Buffer Wait The total time the session spent waiting for buffer wait events. Such events result from a contention in accessing database block buffers; for example, if the requested data block is in an incompatible mode (buffer busy), or if the buffer does not contain sufficient disk space for reading the data block (free buffer). Cache Hit Ratio The effectiveness of the buffer cache, computed by comparing the number of block requests, which found the block in the buffer cache, with the entire block requests (including requests that required disk reads). Chain Rows Ratio This ratio measures the percentage of chained or migrated rows encountered during a fetch as a percentage of the entire fetched rows. • This ratio should be very close to zero, except in applications handling LONG columns. • If this ratio increases over time, usually PCTFREE has been set too low for the LONG containing tables. Client Info The client information; contains the information set by DBMS_APPLICATION_INFO.SET_CLIENT_INFO Client Machine Name The name of the computer where the Client is running Client PID The operating system process ID for the Client program. Cluster Coordination The elapsed time spent while coordinating Real Application Cluster resources at the cluster level. Cluster Wait The total time the session spent waiting for all cluster resources, such as Interconnect hardware, shared disks, and Oracle instances. Consistent Change Ratio The extent to which applications must exercise the read consistency mechanism. The query processing parts of the UPDATE and DELETE operations are subject to read consistency. Consistent Changes The number of times a user process has applied rollback entries to carry out a consistent read on the block. Consistent Reads The number of times a consistent read was requested for a block by this session. Managing Oracle Database Systems: User and Reference Guide 315 Chapter 3—Reference Metric Description Consistent Gets The number of data block reads in CONSISTENT mode (that is, consistent reads). A CONSISTENT read retrieves the block version as of the time the query started. Control File I/O Time spent by a session that issued I/O requests to control file blocks, and is waiting for all of the reads and writes to complete. CPU Usage Time spent waiting for the CPU to process SQL statements and other types of calls. CPU Wait Time spent by the session waiting in the system's run queue to be granted for CPU cycles. Current SQL The last SQL statement executed by the user. Datafile Write Time spent while the DBWR is performing a write to files and blocks (for example, flushing dirty data buffers from the SGA to the data files). DB Block Changes The total number of changes that took part in an update or delete operation made to blocks by this session DB Block Gets The number of data block reads in CURRENT mode (namely, current reads). DB CPU Time Time spent waiting for the CPU to process SQL statements and other types of calls made to Oracle DB Link Wait The time spent waiting for messages, data or break reset to be transmitted over the network interface to another server process on a remote database. DB Time The amount of time used by the database for this session. Dbfile Parallel Read The time spent when the session, which has issued multiple I/O requests in parallel to read blocks from data files into memory, is waiting for all requests to complete. This may occur during recovery or during regular activity, when a session batches many single block I/O requests together and issues them in parallel. Degree of Parallelism The total number of server processes (master and slaves) assigned to handle a single operation. Direct Path I/O Time spent when the session, which has issued asynchronous I/ O requests that bypass the buffer cache, is waiting for them to complete. Direct Path I/O usually reflects disk sorts and bulk loads. Managing Oracle Database Systems: User and Reference Guide 316 Chapter 3—Reference Metric Description Elapsed Time Elapsed time used by this cursor for parsing, executing, and fetching. Enqueue Requests The total number of table or row locks acquired per transaction. Enqueue Waits The total number of times an enqueue request or conversion resulted in a wait. Enqueue Waits Ratio The percentage of enqueue wait events (waits resulting from enqueue requests or conversions) within the total time consumed by enqueue operations. Event The name of the event that resulted in the wait event. External I/O Wait Session waits because of an I/O request on an external large object (LOB), such as a BFILE. Failed Over In a RAC environment, a user can connect to any one of the multiple instances in a cluster. If the connected instance fails, the session can be reconnected automatically to other available instance. This metric indicates whether the session is running in failover mode and failover has occurred (YES) or not (NO). Failover Method The failover method used for determining how failover occurs from the primary node to the backup node Failover Type Failover type settings: SESSION, SELECT, NONE Fixed Table Sequence This metric's value is incremented each time the session completes a call to the database. Lack of changes indicates that the selected session was idle during the selected time range. Free Buffer Wait This wait event occurs when the session needs a free buffer, so it can bring a data block into the buffer cache, and is waiting for a buffer that is not dirty to become available. This can occur if DBWR is not writing dirty buffers to disk fast enough. Full Scan Ratio This ratio measures the percentage of the rows, which were retrieved as a result of full table scans, within the total number of retrieved rows. Managing Oracle Database Systems: User and Reference Guide 317 Chapter 3—Reference Metric Description Global Locks Global Cache Service (GCS) Operations tracks the locations, modes, and roles of data blocks. The GCS therefore also manages the access privileges of (and imposes waits on) various instances in relation to resources. Oracle uses the GCS for cache coherency when the current version of a data block is in one instance's buffer cache and another instance requests that block for modification. If an instance reads a block in exclusive mode, then in subsequent operations multiple transactions within the instance can share access to a set of data blocks without using the GCS. This is only true, however, if the block is not transferred out of the local cache. If the block is transferred out of the local cache, then the GCS updates the Global Resource Directory that the resource has a global role; whether the resource's mode converts from exclusive to another mode depends on how other instances use the resource. Hard Parse Count The number of hard parse requests performed by this session I/O Wait Time spent for the completion of input/output operations. Idle Wait Idle events are events that spend their time idle or waiting for an action. The time keeps incrementing for the event until an action occurs. Kbytes Sent Via SQL*Net To Client Total number of kilobytes sent to the client from the foreground processes Kbytes Sent Via SQL*Net To DBLink Total number of kilobytes sent over a database link Latch Wait Elapsed time spent waiting in latch mode Lock Wait Elapsed time spent waiting in lock mode (enqueue waits) Log Buffer Wait Wait events that occur when copying redo entries into the log buffer. A frequent cause of log buffer wait if redo entries are written into the log buffer faster than LGWR can write it to the redo logs, and user sessions must therefore wait because of log write latency. Log Switch Wait Time spent for the completion of a redo log switch. Log Write Wait Time spent waiting for the completion of write (physical I/O) operation. LGWR experience log write wait events when it periodically flushes the contents of the log buffer into the active log file. Managing Oracle Database Systems: User and Reference Guide 318 Chapter 3—Reference Metric Description Logical Reads The number of logical reads performed by session since login. This value is the total summary of all requests for database blocks, irrespective of whether they were found in the database cache. Logical Writes The number of logical writes performed by the session since login. Logon Time The date and time when the user logged on to the database. The format is year/month/day and hours:minutes:seconds (on a 24 hour clock). Logout Time The date and time when the user logged out of the database. The format is year/month/day and hours:minutes:seconds (on a 24 hour clock). Miscellaneous I/O Sporadic I/O activity, such as opening a data file. This wait event is usually negligible. Miscellaneous Other Time spent waiting for miscellaneous operations to complete. None of the operations can be classified into any other wait categories. Module Name The currently executing module name, set by DBMS_APPLICATION_INFO.SET_MODULE. Multi Block Read Time spent waiting for the session to read multiple data blocks from the data files into the SGA buffer. Multi-Block Read waits usually reflects full table or index scans. Generally full scans of large segments should be avoided, although they are sometimes unavoidable, as in data warehouse environments. Network Wait Time spent waiting for messages and data to be sent or received over the network interface. Network/Interconnect Wait events caused by the communication link between the nodes. Non Buffer Activities The time spent on active wait events, excluding buffer wait. Non Cluster Activities The time spent on active wait events, excluding cluster wait. Non CPU Activities The time spent on active wait events, excluding CPU wait. Non I/O Activities The time spent on active wait events, excluding I/O wait. Non Latch Activities The time spent on active wait events, excluding latch wait. Managing Oracle Database Systems: User and Reference Guide 319 Chapter 3—Reference Metric Description Non Lock Activities The time spent on active wait events, excluding lock wait. Non Network Activities The time spent on active wait events, excluding network wait. Non Other Activities The time spent on active wait events, excluding other wait. Non Redo Activities The time spent on active wait events, excluding redo wait. Oracle User Name The Oracle database account of the started session. OS User Name The operating system user used by the client Other DB Buffer Wait Sporadic buffer wait events, caused by buffer contention, such as buffer write wait. Usually negligible. Other Wait Time spent waiting for miscellaneous operations to complete. None of these operations fits into the other wait categories. Parallel Status The Parallel Query status; can have one of the following values: • ENABLED — the session is in a Parallel Query enabled mode • DISABLED — the Parallel Query enabled mode is not supported for the session • FORCED — the session has been altered to force Parallel Query) Parse Count The number of parse requests performed by this session. Physical Reads The number of data blocks read by Oracle, which are not present in the buffer cache or in the PGA, thereby requiring I/O read operations, that is: the total number of data blocks read from disk. Physical Reads Direct Number of reads that were carried out directly from disk into the PGA, bypassing the buffer cache. Physical Writes The number of physical writes carried out by the session since logon Physical Writes Direct Number of writes directly to disk, bypassing the buffer cache Recursive Calls Number of Recursive Calls generated at both the user and system level. Redo Entries The number of redo entries issues by the session since logon Managing Oracle Database Systems: User and Reference Guide 320 Chapter 3—Reference Metric Description Redo Log Space Requests Number of times the active log file is full and the session must wait for a log switch to be performed. After the log switch completes, the redo log entries can be allocated, and the session resumed. Log files that are small in relation to the size of the SGA or the commit rate of the workload can cause performance problems. Redo Log Space Wait Ratio This ratio measures the redo memory allocation. If it is greater than 1/5000, then the redo log buffer should be increased until the redo log space wait ratio stops decreasing. This ratio is calculated as follows: Redo Log Space Requests Redo Entries Redo Wait Elapsed redo wait time, including the time spent waiting on contention resulting from: • User sessions writing redo entries to the redo log buffer • LGWR writing the log buffer contents into the active redo log file • ARCH archiving redo logs to the destination location Redo Writes Total number of write operations issued by LGWR to the redo log files. Resource Consumer Group Name The name of the session's current resource consumer group. Resource Manager Waits experienced by sessions when exceeding thresholds/ quotas set for each session by the resource manager. Roundtrips To From Client Total number of *Net messages sent to and received from the Client Rows Processed Total number of rows the parsed SQL statement returns. Schema Name The name of the database schema. Seconds in Wait The wait event’s length of time, in seconds, since logon Serial Number The session’s serial number. The combination of SID and serial number provides a unique identification of the session. Managing Oracle Database Systems: User and Reference Guide 321 Chapter 3—Reference Metric Description Server Indicates the server type. The possible values are as follows: • DEDICATED • SHARED • PSEUDO • POOLED • NONE Server PID The operating system process ID for the Oracle shadow process. Server Program Name The name of the application, as displayed by the OS running it (for example: sqlplus.exe). Service Name The service name of the session. Session Cache Hits Number of hits in the session cache. A hit means that the SQL statement did not have to be re-parsed. Session Cursor Cache Hits Number of hits in the session cursor cache. A hit means that the SQL statement did not have to be re-parsed. Session ID The session's identification number Session PGA Memory The Amount of PGA memory allocated for the current session. Session SQL ID The session identification number for the SQL statement. Session Status The session status; indicates whether the session is currently executing SQL statements (ACTIVE) or not (INACTIVE). Session Type The session type (USER or BACKGROUND). Session UGA Memory The amount of UGA memory allocated to the current session. Session Up Time The amount of time the session was running (up) during the specified time range. Single Block Read Number of read requests that read one database block per request. Sort Disk Total number of Sort operations carried out by the session, which required creating temporary segments on disk in order to store the intermediate sort results. Such a need may indicate that the Sort operations overloaded the PGA memory. Managing Oracle Database Systems: User and Reference Guide 322 Chapter 3—Reference Metric Description Sort Overflow Ratio This ratio measures the portion of sorts that could not fit into a session's sort area and therefore used temporary segments. High values may indicate a need to increase the value of SORT_AREA_SIZE parameter (in the initialization files), in order to allow more memory space for sort operations. However, if the application requires sorting of vast number of rows, the use of disk sorting and temporary segments may be unavoidable. The ratio is calculated as follows: Sorts to Disk/(Sorts to Disk + Sorts in Memory) Sort Rows The total number of rows sorted. Sorts Disk Sorts that require creation of temporary segments on disk to store the intermediate sort results. Such sorts are carried out if the data being sorted cannot be fit into memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. Sorts In Memory Sort operations that were carried out entirely in memory, without requiring any disk writes. The sort operation can take place in memory if the sorted data can be fit into the memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. SQL Executions Number of statements sampled during the specified time range. SQL Executions Ended Number of SQL statements whose running ended during the specified time range. SQL Executions Started Number of SQL statements that started during the specified time range. SQL*Net Time spent waiting for messages, data or break reset to be transmitted over the network interface to the Client. SQL*Net Roundtrips To/ From Client Total number of *Net messages sent to and received from the Client Start Time The time when the session started. Table Fetch By ROWID Number of rows that were fetched using a ROWID (usually recovered from an index). Table Fetch Continued Row Number of times a chained or migrated row is encountered during a fetch. Table Scans Long Tables The number of times Oracle carried out a complete scan of a long table Managing Oracle Database Systems: User and Reference Guide 323 Chapter 3—Reference Metric Description Table Scans ROWID Ranges The number of table scans conducted during parallel query by the query slaves, with the specified ROWID ranges Table Scans Rows Gotten The number of rows scanned during full table scans. High values may indicate use of non-optimal queries, absence of proper indexing, or both. Table Scans Short Tables The number of times Oracle carried out a complete scan of a short table. Transaction Rollback Ratio The percentage of transactions that had to be rolled back out of the entire number of transactions that ran during the specified time range. User Calls Number of user calls such as login, parse, fetch, or execute. This metric value represents the rate at which the user application addresses Oracle. User Calls Per Parse The average number of calls in relation to statement parses. Changes in this rate most likely result from application change, but may also indicate that usage patterns are changing and users are moving from one module to another, either more or less frequently. User Commits Number of transactions committed during the current interval. After a transaction is committed, the redo generated by that transaction reflects the changes made to database blocks and they must be written to disk. User Rollbacks Number of transactions that were rolled back. Rollbacks occur either because of a manually issuing a ROLLBACK command or in the event of an error Wait Class The classification of wait events. Sessions Summary Purpose The Sessions Summary collection retrieves information regarding session types. Collection Type Oracle Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions Metric Description Active Sessions The average number of active sessions during the specified time range Active Sessions PCT The average percentage of active sessions within the total sessions during the specified time range Background Sessions The average number of background sessions that ran during the specified time range. Blocked Sessions The average number of blocked processes that were collected during the specified last sample. Cached Sessions The average number of cached sessions during the specified time range Foreground Sessions The average number of foreground sessions that ran during the specified time range. Inactive Sessions The average number of inactive sessions during the specified time range Killed Sessions The average number of killed sessions during the specified time range PDB In multitenant architectures, the name of the pluggable database Sniped Sessions The average number of sniped sessions during the specified time range Total Sessions The average number of sessions during the specified time range Total Sessions Limit The maximum allowable number of sessions 324 Managing Oracle Database Systems: User and Reference Guide 325 Chapter 3—Reference Servers Purpose The Servers collection provides instance processes. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Dedicated Servers The number of dedicated server processes. Each dedicated server process service a single client process. Dispatchers The number of multi-threaded server (MTS) dispatchers. MTS dispatchers coordinate the allocation of shared servers to client tasks. Shared Servers The number of servers that service more than one client process. This value includes both shared servers and multithreaded servers (MTS). Subcategories Wait Purpose The Subcategories Wait collection provides instance wait events grouped into resource categories. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Frequency Mode Collection Interval (Seconds) Online 60 Offline 300 Metric Descriptions Metric Description Archival Wait Any activity that generates an LGWR (log writer) for archiving the redo logs to the target location. Additionally, user sessions can also experience archival waits when the login to which the LGWR is going to switch has not yet been archived and the session is issuing DML statements. Backup/Recovery Wait When archiving is not performed fast enough, database operations can be suspended temporarily whenever the log writer is forced to wait for an inactive group to become available for reuse. Balance Cluster Wait The extent to which the selected instance shares the balance of cluster wait events across all RAC nodes (as a percentage). Ideally, all instances should share the cluster wait time equally; however, this metric can indicate a situation where a specific instance takes a much larger share, thereby possibly overloading the system. Balance I/O Wait The extent to which the selected instance shares the balance of I/O wait events across all RAC nodes (as a percentage). Ideally, all instances should share the I/O wait time equally; however, this metric can indicate a situation where a specific instance takes a much larger share, thereby possibly overloading the system. Balance Workload The extent, in percent, to which workload (average active sessions) is balanced across all RAC nodes. Ideally, all instances should equally share the wait time caused by active sessions workload; however, this metric can indicate a situation where a specific instance takes a much larger share, thereby possibly overloading the system. Buffer Busy Wait The time spent by the session when trying to access a data block that is unavailable for one of the following reasons: • The data block is currently not in memory, but another process has already issued an I/O request to read it into memory • The data block is in memory, but in an incompatible mode (for example, current instead of consistent) 326 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Cluster Coordination Wait The elapsed time spent while coordinating Real Application Cluster resources at the cluster level. Cluster Wait The total wait for all cluster resources, such as Interconnect hardware, shared disks, and Oracle instances. Cluster Wait Percent The percentage of cluster wait events for the overall active time. Control File IO Wait Time spent by a session that issued I/O requests to control file blocks, and is waiting for all of the reads and writes to complete. CPU Usage Wait Time spent waiting for the CPU to process SQL statements and other types of calls. CPU Wait Wait Time spent by the session waiting in the system's run queue to be granted for CPU cycles. Datafile Write Wait Time spent while the DBWR is performing a write to files and blocks (for example, flushing dirty data buffers from the SGA to the data files). DB Buffer Wait The time spent waiting for the completion of buffer operations. This wait event usually results from insufficient or non-optimal buffer configuration. DB File Parallel Read Wait The session issued multiple I/O requests in parallel. As these requests require reading blocks from data files into memory, the session is waiting for all requests to complete. This may occur during recovery or during regular activity, when a session batches many single block I/O requests together and issues them in parallel. DB Link Wait The time spent waiting for messages, data or break reset to be transmitted over the network interface to another server process on a remote database. Direct Path IO Wait Time spent when the session, which has issued asynchronous I/ O requests that bypass the buffer cache, is waiting for them to complete. Direct Path I/O usually reflects disk sorts and bulk loads. Enqueue Wait Percent The percentage of time spent waiting on enqueue out of the total active time. External File IO Wait Time spent waiting for the completion of an I/O request on an external large object (LOB), such as a BFILE. 327 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Flash Cache Single Block Read Waits Percent The percentage of time, within the total active time, spent carrying out single block reads from flash cache. Flash Cache: Write Complete Waits Time spent waiting for the DBWR to write a block from the buffer cache to the flash cache. Flash Cache: Write The percentage of time, within the total active time, spent Complete Waits Percent waiting for the DBWR to write a block from the buffer cache to the flash cache. Free Buffer Wait The session needs a free buffer, so it can bring a data block into the buffer cache, and is waiting for a buffer that is not dirty to become available. This can occur if DBWR is not writing dirty buffers to disk fast enough. Global Locks Wait Global Cache Service (GCS) Operations tracks the locations, modes, and roles of data blocks. The GCS therefore also manages the access privileges of (and imposes waits on) various instances in relation to resources. Oracle uses the GCS for cache coherency when the current version of a data block is in one instance's buffer cache and another instance requests that block for modification. If an instance reads a block in exclusive mode, then in subsequent operations multiple transactions within the instance can share access to a set of data blocks without using the GCS. This is only true, however, if the block is not transferred out of the local cache. If the block is transferred out of the local cache, then the GCS updates the Global Resource Directory that the resource has a global role; whether the resource's mode converts from exclusive to another mode depends on how other instances use the resource. Idle Wait Idle wait events are events that spend their time idle or waiting for an action. The time keeps incrementing for the event until an action occurs. IO Wait Time spent for the completion of input/output operations Latch Wait Elapsed time spent waiting in latch mode Lock Wait Elapsed time spent waiting in lock mode (enqueue waits) 328 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Log Buffer Wait Wait events that occur when copying redo entries into the log buffer. A frequent cause of log buffer wait if redo entries are written into the log buffer faster than LGWR can write it to the redo logs, and user sessions must therefore wait because of log write latency. Log Switch Time Percent The percentage of log switch wait events within the overall active time. Log Switch Wait Time spent for the completion of a redo log switch. Log Write Wait Time spent waiting for the completion of write (physical I/O) operation. LGWR experience log write wait events when it periodically flushes the contents of the log buffer into the active log file. Miscellaneous IO Wait Sporadic I/O activity, such as opening a data file. This wait event is usually negligible. Multiblock Read Wait The session is reading multiple data blocks from the data files into the SGA buffer. Multi-Block Read wait events usually reflect full table or index scans. Generally full scans of large segments should be avoided, although they are sometimes unavoidable, as in data warehouse environments. Network Wait Time spent waiting for messages and data to be sent or received over the network interface. Network/Interconnect Wait Wait events caused by the communication link between the nodes. Non Buffer Wait The time spent on active wait events, excluding buffer wait. Non Cluster Wait The time spent on active wait events, excluding cluster wait events. Non IO Wait The time spent on active wait events, excluding I/O wait events. Non Latch Wait The time spent on active wait events, excluding latch wait events. Non Lock Wait The time spent on active wait events, excluding lock wait events. Non Network Wait The time spent on active wait events, excluding network wait events. Non Redo Wait The time spent on active wait events, excluding redo wait events. 329 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Other DB Buffer Wait The sporadic buffer waits caused by buffer contention, such as buffer write wait. This wait event type is usually negligible. Other Other Wait Assorted independent waits. Usually negligible. Other Wait Time spent waiting for the completion of miscellaneous operations, namely: operations that do not fit into the other wait categories. Redo Wait The elapsed time, including the contention, resulting from the various redo operations: • User sessions writing redo entries to the redo log buffer • LGWR (log writer) writing the log buffer contents into the active redo log file • ARCH archiving redo logs to the destination location Resource Manager Wait The waits experienced by sessions when exceeding thresholds or quotas permitted by the resource manager Single Block Read Wait The session is reading a single block from a datafile into the SGA buffer SQL Net Wait The wait time it takes for messages, data, or break reset to be transmitted over the network interface to the client Total Wait Sum of all active wait events. The value of this metric equals the value of the session’s total activity within the given time period. Tablespaces Free Space Fragmentation Purpose The Tablespaces Free Space Fragmentation collection retrieves information regarding the fragmentation of tablespace free space. Collection Type Oracle Collection Sampling Settings None (on-demand collection). 330 Managing Oracle Database Systems: User and Reference Guide 331 Chapter 3—Reference Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Average Extent Size The average size of free extents in this tablespace Extents The number of free extents in this tablespace Max Extent Size The size of the largest free extent in the selected tablespace Min Extent Size The size of the smallest free extent in this tablespace Partly Usable The total sum of fragments that are smaller than the maximum allocated extent in the tablespace, but bigger than, or equal to, the minimum allocated extent in the tablespace PDB In multitenant architectures, the name of the pluggable database Tablespace Name The name of the currently selected tablespace Unusable Total sum of fragments whose size is less than the size of the minimum extent in the tablespace Usable Total sum of fragments that are greater than or equal to the minimum extent in the tablespace. Tablespaces Storage Purpose The Tablespaces Storage collection provides a list of all tablespaces and their overall properties and storage characteristics. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 3600 Online 3600 Offline 14400 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Allocated Space Used Pct The percent of used allocated space within this tablespace Allocation Type The type of extent allocation to be used for this tablespace. The possible values for this metric are as follows: • USER • SYSTEM • UNIFORM Auto DF ASM The number of autoextensible ASM datafiles Auto DF FS The number of autoextensible OS datafiles Average Fragment Size The average size of fragments in this tablespace Big File Indicates whether the tablespace is a BIGFILE tablespace. BIGFILE tablespaces are supported only for locally managed tablespaces with automatic segment-space management. Block Size The tablespace blocksize Contents The type of tablespace contents is one of the following: • PERMANENT • TEMPORARY • UNDO Database The database name Default Table Compression Indicates whether table compression is enabled by default for this tablespace Encryption Indicates whether the tablespace data is encrypted. Estimated Date Until Full The estimated date when the tablespace becomes full. Estimated Until Full The estimated number of days until the tablespace is full. Extent Management Indicates the extent management type: • DICTIONARY — where the management of extents is within the Oracle data dictionary. • LOCAL — where the management of extents is within the tablespace itself. 332 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Force Logging Indicates whether the tablespace is in FORCE LOGGING mode: • Yes — logs all changes to all objects in the tablespace except changes to temporary segments. • No — logs are not created. Fragments The number of fragments in this tablespace Free The total amount of free space in this tablespace Free Pct The percent of free space in this tablespace Logging Indicates whether a redo log is to be generated for the tablespace Long Term Growth Rate The long term growth rate. Manual DF ASM The number of manual ASM datafiles Manual DF FS The number of manual operating system datafiles Max Fragment Size The size of the largest fragment in this tablespace Min Fragment Size The size of the smallest fragment in this tablespace Number of Datafiles The number of datafiles in the tablespace PDB In multitenant architectures, the name of the pluggable database Recycle Free Pct The percentage of recycle bin space which occupies the allocated free space of the tablespace. Recycle Free The total amount of space used by the recycle bin within the allocated free space of the tablespace Retention Indicates whether unexpired UNDO operations are retained in the tablespace. This applies only to UNDO tablespaces. Segment Space Management Indicates the method used for managing free and used segments: • manually — by free lists • automatically — by bitmaps Short Term Growth Rate The short term growth rate. 333 Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Metric Description Status The current status of the tablespace. • ONLINE • OFFLINE Tablespace Name The name of the currently selected tablespace. Total The total amount of space in the currently selected tablespace. Used The amount (in megabytes) of space used in this tablespace Used Growth The amount (in megabytes) by which the size of the tablespace has increased during the specified time range Top Blocked Objects Purpose The Top Blocked Objects collection provides a list of top blocked objects for TX locks. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 60 Offline 300 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Segment Name The name of the segment Segment Waits The number of waits for a specific segment during the selected time range 334 Managing Oracle Database Systems: User and Reference Guide 335 Chapter 3—Reference Top Sessions Purpose The Top Sessions collection provides instance current sessions details. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 20 seconds Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Action The currently executing action name set by DBMS_APPLICATION_INFO.SET_MODULE BG CPU Time The amount of CPU time used by a session from the time a user call starts until it ends BG Time The amount of time used by a session from the time a user call starts until it ends Blocking Session Status The status of the blocking session; indicates whether the session is currently executing SQL statements (ACTIVE) or not (INACTIVE). Client Info The client information; contains the information set by DBMS_APPLICATION_INFO.SET_CLIENT_INFO Client Machine Name The name of the computer where the Client is running Client PID The operating system process ID for the Client program Consistent Changes The number of times a user process has applied rollback entries to carry out a consistent read on the block Consistent Gets The number of times a consistent read was requested for a block by this session Current SQL The last SQL statement executed by the user DB Block Changes The total number of changes that took part in an Update or Delete operation made to blocks by this session DB Block Gets The number of times a current block was requested by this session Managing Oracle Database Systems: User and Reference Guide 336 Chapter 3—Reference Metric Description DB CPU Time Time spent waiting for the CPU to process SQL statements and other types of calls made to Oracle DB Time The amount of time used by the database for this session Event The name of the event that resulted in the wait event Failed Over In a RAC environment, a user can connect to any one of the multiple instances in a cluster. If the connected instance fails, the session can be reconnected automatically to other available instance. This metric indicates whether the session is running in failover mode and failover has occurred (YES) or not (NO). Failover Method The failover method used for determining how failover occurs from the primary node to the backup node Failover Type Failover type settings: SESSION, SELECT, NONE Fixed Table Sequence This metric's value is incremented each time the session completes a call to the database. Lack of changes in this metric's value indicates that the selected session was idle during the selected time range. Hard Parse Count The number of hard parse requests performed by this session Logical Reads The number of logical reads performed by session since login. This value is the total summary of all requests for database blocks, irrespective of whether they were found in the database cache. Logical Writes The number of logical writes performed by the session since login. Logon Time The date and time when the user logged on to the database. The format is year/month/day and hours:minutes:seconds (on a 24 hour clock). Module Name The currently executing module name, set by DBMS_APPLICATION_INFO.SET_MODULE. Oracle User Name The Oracle database account of the started session OS User Name The operating system user used by the client Parallel Status The Parallel Query status; can have one of the following values: • ENABLED — the session is in a Parallel Query enabled mode • DISABLED — the Parallel Query enabled mode is not supported for the session • FORCED — the session has been altered to force Parallel Query) PDB In multitenant architectures, the name of the pluggable database Managing Oracle Database Systems: User and Reference Guide 337 Chapter 3—Reference Metric Description Physical Reads The number of data blocks read by Oracle, which are not present in the buffer cache or in the PGA, thereby requiring I/O read operations, that is: the total number of data blocks read from disk. Physical Writes The number of physical writes carried out by the session since logon Redo Entries The amount of redo entries issues by the session since logon Resource Consumer Group Name The name of the session's current resource consumer group Schema Name The name of the database schema Seconds in Wait The wait event’s length of time, in seconds, since logon Serial Number The session’s serial number. The combination of SID and serial number provides a unique identification of the session. Server The server’s type: DEDICATED, SHARED, PSEUDO. or NONE Server PID The operating system process ID for the Oracle shadow process. Server Program Name The server program name Service Name The service name of the session Session Action The currently executing action name, set by DBMS_APPLICATION_INFO.SET_MODULE. Session Cursor Cache Hits The number of session cursor cache hits. Session ID The session identification number. Session PGA Memory The amount of PGA memory allocated for the current session Session SQL ID The session identification number for the SQL statement Session Status The session status; indicates whether the session is currently executing SQL statements (ACTIVE) or not (INACTIVE) Session Type The session type (USER or BACKGROUND) Session UGA Memory The amount of UGA memory allocated to the current session Sort Disk The number of sort operations that require creation of temporary segments on disk to store the intermediate sort results. This occurs if the data being sorted cannot be fit into memory block specified by the SORT_AREA_SIZE parameter in the init.ora file. Managing Oracle Database Systems: User and Reference Guide 338 Chapter 3—Reference Metric Description Wait Class The classification of wait events. Top SQLs Purpose The Top SQLs collection retrieves top n SQL statements (namely: statements that experienced the longest time of CPU consumption or total wait events during the specified time range) and their details. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Offline 900 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Array Wastage The gap (by a percentage) between the average number of rows per fetch that the array can hold, as defined in the arraysize parameter of the various fetch operations of the specified SQL query, and the number of rows actually retrieved. For example: if the average setting of the arraysize parameter was 5, and the average number of rows actually retrieved per operation was 10, the value indicated will be 50. Ideally, the value of this parameter is 0; any other value indicates the need for additional sql.net roundtrips per each SQL query execution. Buffer Gets Number of buffer gets for this child cursor Child Number Hash value of the parent statement in the library cache CPU Time CPU time used by this cursor for parsing, executing, and fetching Direct Writes Number of direct writes for this child cursor Disk Reads Number of disk reads for this child cursor Managing Oracle Database Systems: User and Reference Guide 339 Chapter 3—Reference Metric Description Elapsed Time Elapsed time used by this cursor for parsing, executing, and fetching Executions Number of executions that took place on this object since it was brought into the library cache Fetches Number of fetches associated with the SQL statement Fetches per Execution The average number of fetches per a single execution associated with the SQL statement First Load Time The first time the object was loaded to the shared pool Hash Value Hash value of the parent statement in the library cache Java Time Java execution time Loads Number of times the object was either loaded or reloaded Loads per Execution The average number of loads per a single execution associated with the SQL statement Other Buffer Gets Total time spent waiting for buffer gets on all SQL statements other than the selected statement. Other CPU Time Total CPU time consumed for executing all SQL statements other than the selected statement. Other Disk Reads The total number of disk reads carried out by all SQL statements except the selected statement. Other Executions Number of executions that took place on this object since it was brought into the library cache. Other Total Waits Total time spent on wait events due to executing all SQL statements except the selected statement. Parallel Executions Total number of executions performed by Parallel execution servers. Parse Calls Number of parse calls for this child cursor. PLSQL Time PL/SQL execution time. Rows Processed Total number of rows the parsed SQL statement returns Sorts Number of sorts that were done for the child cursor Sorts per Execution The average number of sorts per a single execution associated with the SQL statement Managing Oracle Database Systems: User and Reference Guide 340 Chapter 3—Reference Metric Description Total Waits The total time spent on wait events. Top SQLs Full Text Purpose This collection retrieves the SQL statement’s full text. Collection Type Oracle Collection Sampling Settings The Top SQLs Full Text collection is an on-demand collection, which retrieves data only upon clicking View full text. Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Hash Value The SQL statement’s hash value SQL Full Text Full text for the SQL statement exposed as a Character Large OBjects (CLOB) column Top SQLs Text Purpose The Top SQLs Text collection retrieves the first n characters of the skoal statement text (the exact number of the characters is defined in the Top SQL Statements pane of the Collection Settings screen). Collection Type Oracle Collection Sampling Settings This collection retrieves SQL text only for new hash values. Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Hash Value The SQL statement’s hash value. Managing Oracle Database Systems: User and Reference Guide 341 Chapter 3—Reference Metric Description Is Full Text A boolean value that indicates whether the first n characters are a full or partial SQL text for the current cursor. SQL Text First n characters of the SQL text for the current cursor. Total Archive Storage Purpose The Total Archive Storage collection collects disk space data for critical archive destinations. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 900 Offline 3600 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Archive Destination Free Space The free space size (in megabytes) in the file system where the most critical archive destination resides. Archive Destination Size The total size (in megabytes) of the file system where the most critical archive destination resides. Critical Archive Time To Failure The number of hours remaining until the archive destination with the shortest time to failure is filled. Destination The destination ID of the most critical archive destination. Total Database Storage Purpose The Total Database Storage collection retrieves the database’s total storage, that is: the summarized value of total free space and total allocated space. Managing Oracle Database Systems: User and Reference Guide 342 Chapter 3—Reference Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 300 Online 900 Offline 3600 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Maximum Used Percent The maximum percent of the used disk space of all the database datafiles. Total Allocated Space The total size (in megabytes) of allocated space in the database. Total Allocated Space Delta The incremental change in the total size of the database’s allocated space during the specified time range. Total Allocated Space Rate The rate (in megabytes/seconds) at which disk space was allocated on the database during the last sample. Total Free Space Percent The percent of the total free disk space within the database. Total Instance Waits Purpose The Total Instance Waits collection collects the total amount of time (in seconds) that the instance spent on wait events since the last sample. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 300 Managing Oracle Database Systems: User and Reference Guide 343 Chapter 3—Reference Frequency Mode Collection Interval (Seconds) Offline 900 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Total Waits The total amount of time (in seconds) that the instance spent on wait events since the last sample. Undo Activity Purpose The Undo Activity collection collects UNDO operations-related information. Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 600 Online 600 Offline 600 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Active UNDO Tablespace The name of the last active undo tablespace of the last period when undo statistics were collected (by default: every 10 minutes) Begin Time The beginning of the last period when undo statistics were collected End Time The end of the last period when undo statistics were collected Expired Blocks Removed Number of expired undo blocks stolen from other undo segments Managing Oracle Database Systems: User and Reference Guide 344 Chapter 3—Reference Metric Description Expired Blocks Reused Number of expired undo blocks reused within the same undo segments Expired Blocks Steals Number of attempts to steal expired undo blocks from other undo segments Highest Concurrently TNXs The highest number of transactions executed concurrently within the specified time range Longest Query The length, in seconds, of the longest query executed in the instance during the specified time range Longest Query SQL ID SQL identifier of the longest running SQL statement within the specified time range No Space Error Count The number of times requests for space in the undo tablespace failed to find any free space available SnapShot Too Old Error Count The number of times the error ORA-01555 (snapshot too old) occurred Transactions The total number of transactions executed within the specified time range Unexpired Blocks Removed Number of unexpired blocks removed from certain undo segments so they can be used by other transactions Unexpired Blocks Reused Number of unexpired undo blocks reused by transactions Unexpired Blocks Steals Number of attempts to obtain undo space by stealing unexpired extents from other transactions Used Blocks The total number of undo blocks consumed Total Summary Purpose The Total Summary collection collects UNDO definitions, storage and retention. Collection Type Oracle Managing Oracle Database Systems: User and Reference Guide 345 Chapter 3—Reference Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 600 Online 600 Offline 600 Metric Descriptions \\\\\\\\\\\\\\\\\\\\\\\\\\\ Metric Description Active Undo Tablespace Name of the last active undo tablespace Actual Undo Size The size (in megabytes) of the undo tablespace Autoextensible Indicates whether the datafile is autoextensible Current Retention Specifies (in seconds) the low threshold value of undo retention Minimum Required Undo Size The minimum size (in megabytes) of undo space required. This value takes into account transactions rate, undo generation rate and current undo retention. Optimal Retention The optimal undo retention, taking into account undo generation rate and actual undo size Retention Guarantee Indicates the state of the Undo tablespace retention (GUARANTEE/NOGUARANTEE) Retention value of GUARANTEE indicates that unexpired undo, in all undo segments in the undo tablespace, should be retained even at the cost of failure of forward-going operations that need to generate undo in those segments. Transactions Count The rate of transaction executed within the specified time range Undo Generated Volume The rate per second (in megabytes) of undo generated within the specified time range Undo Management Indicates whether undo management is automatic or manual Usability Purpose The Usability collection provides retrieved instance usability based on response time and instance availability. Managing Oracle Database Systems: User and Reference Guide Chapter 3—Reference Collection Type Oracle Collection Sampling Settings Frequency Mode Collection Interval (Seconds) Realtime Collection 60 Online 60 Offline 300 Metric Descriptions Metric Description Availability The percentage of time, within the selected time range, during which the database was available: • For a RAC node — the percentage of nodes that are up and running • For a single instance — the percentage of the time during this period the instance was up and running Connect Availability The percentage of time, within the selected time range, during which the connection to the database instance was successful. Connection Time The amount of time it takes to establish a connection. Error Message Last error message that was received upon trying to connect to the database (if an unsuccessful connection trial took place). Listener Error Message The last error message (if any) for each listener Listener Name The listener name for each listener Listener Status For each listener: the percentage of time, within the selected time range, during which the listener was available. Important Retrieving the Listener Status information requires having an OS user with the permission to run the lsnrctl command. OS Connect Availability The percentage of time, within the selected time range, during which the connection to the OS was successful. OS Error Message The last error message (if any) received upon an unsuccessful connection trial. 346 Managing Oracle Database Systems: User and Reference Guide 347 Chapter 3—Reference Metric Description Response Time The amount of time (in microseconds) used by the active cursor for carrying out parsing/executing/fetching operations Up Since The date and time when the instance started Glossary Buffer Wait Buffer wait events result from a contention in accessing database block buffers; for example, if the requested data block is in an incompatible mode (buffer busy), or if the buffer does not contain sufficient disk space for reading the data block (free buffer). CPU Usage When SQL statements and other types of calls are made to Oracle, processing the calls requires spending a certain amount of CPU time. Whereas processing average calls requires a small amount of CPU time, a SQL statement involving a large amount of data or a runaway query can consume a much larger amount, thereby reducing the CPU time available for other processing. CPU utilization is a key operating system statistic in the tuning process. Excessive CPU usage can result from an inadequately-sized system, untuned SQL statements, or inefficient application programs. CPU Wait CPU wait events take place when the session is waiting in the system's run queue to be granted for CPU cycles. The length of these wait events (the amount of time spent) depends upon the number of concurrent processes and threads requesting CPU time. This metric’s value should be inspected in conjunction with the value of the Run Queue Length. Cluster Wait Cluster wait events take place when the session waits for all types of cluster resources, such as Interconnect hardware, shared disks, and Oracle instances. I/O Wait I/O wait events take place when the session is waiting for the completion of input/output operations. The length of these wait events (the amount of time spent) depends upon the number of concurrent processes and threads requesting CPU time. This metric’s value should be inspected in conjunction with the value of the Run Queue Length. Managing Oracle Database Systems: User and Reference Guide 348 Chapter 3—Reference Latch Wait Latch wait events occur when a session needs to acquire a latch that is held by another session. Latches are simple, low-level serialization mechanisms to protect shared SGA data structures and shared code segments from simultaneous session access. They are similar in purpose to locks: latches protect internal memory structures while locks protect data structures. Latch wait events are designed to be very quickly acquired and freed. Lock Wait Lock wait events occur when a session attempts to acquire a lock that is already held by another session. A session will be blocked until the blocking session releases the lock. Locks are designed to ensure data integrity by limiting simultaneous data access. Multi-user database locking generally consists of two levels: exclusive locks and share locks. You want to watch out for exclusive locks (that is, TX) as they prohibit resource sharing. For example, the first transaction that exclusively locks a resource is the only one that can alter the resource (except for the DBA) until the exclusive lock is released. Share locks, unlike exclusive locks, allow a resource to be shared. Most locking issues are application-specific and can be addressed by tuning the concurrency logic in the application. Network Wait Network wait events occur when a session spends time waiting for messages to be sent or received over the network interface. Network performance, which is measured in number (per second) of packets sent and received, can be used just like disk statistics to detect overload or non-optimal performance within a network or a network interface. For more information, see “SQL*Net” on page 349. Excessive network wait can result from either: • Excessive network usage, originating in the application • Physical issues, identifiable by network errors and network collisions Other Wait Other wait events refer to time spent waiting for miscellaneous operations to complete. None of these operations fits into the separately identified wait event. Recursive Calls When Oracle needs to make a change to the tables maintained for internal processing, it internally generates an internal SQL statement, which in turn generates a recursive call. Recursive calls occur because of cache misses and dynamic storage extension. If the dictionary data is found in cache, a recursive call is not made and the data is read from cache directly. Managing Oracle Database Systems: User and Reference Guide 349 Chapter 3—Reference SQL*Net SQL *net is the Oracle network interface that allows Oracle tools running on network workstations and servers to access, modify, share, and store data on other servers. SQL *net is considered part of the program interface in network communications. Managing Oracle Database Systems: User and Reference Guide 350 Index A activity dashboard 122 activity locks panel 125, 128 adding and configuring agents 10 adding ASM instances 34 agent DB_Oracle 161, 162, 164 DB_Oracle_RAC 162 DB_Oracle_RAC_Instance 163 agent properties DB_Oracle 162 DB_Oracle_RAC 163 DB_Oracle_RAC_Instance 163 agents adding and configuring 10 alarms See also email notifications customizing 175 customizing alarm messages 181 editing thresholds and severity levels 179 enabling 179 monitoring 99 sensitivity level 176 alert log administration screen ignore list 191 match list 191 alert log messages chart 156 alert log messages table 157 alert log panel 156 architecture Foglight for Oracle 21 Archive Destination Miscellaneous collection 226 archive log destinations table 148 archive logs panel 148 Archive Logs Storage collection 227 archive logs storage collection 227 ASM Diskgroup Balance collection 231 ASM Diskgroups collection 229 ASM Disks collection 231 ASM Instance Total collection 233 ASM instances adding 34 ASM Operations collection 234 auto discovery overview 10 wizard 24 B Backup Config collection 235 Backup Jobs collection 236 Backup Type Status collection 237 baseline chart format 57 baseline charts 57 bounds editing for alarms 179 breakdown chart format 57 breakdown charts 57 C cell offload efficiency 159 cell performance 159 Change Tracking collection 237 chart baseline 57 breakdown 57 chart formats baseline 57 breakdown 57 Cluster Latency collection 238 cluster usability panel 124 clustered environment 24 Clusterware modifying the monitoring settings 41 monitoring through CRS agents 39 collection Archive Destination Miscellaneous 226 Archive Logs Storage 227 ASM Diskgroup Balance 231 ASM Diskgroups 229 Managing Oracle Database Systems: User and Reference Guide 351 Index ASM Disks 231 ASM Instance Total 233 ASM Operations 234 Backup Config 235 Backup Jobs 236 Backup Type Status 237 Change Tracking 237 Cluster Latency 238 Cluster Misses Cluster Misses collection 242 Event Wait 251 Filesystem Disk 259 Flashback Area 260 History Locks Tree Instance 261 History Locks Tree RAC 263 Host Properties 265 Initialization Parameters 266 Instance General 268 Invalid Objects 268 Invalid Objects List 270 IO Activity 270 Jobs Queue 277 Latches Activity 278 Locks Activity 279 Locks Breakdown 280 Locks_Instance 281 OS and instance statistics 285 OS General Network Summary 293 PA Instance Statistics 293 PA Usability 296 Parallel 297 Pluggable Database State 301 Redo Instances Misc 303 Session Details Datafile I/O PA 305 Session Details Locks PA 306 Session Details Open Cursors PA 307 Session Details SQL Summary PA 308 Session Details Statistics PA 309 Session List PA 312 Sessions Summary 323 Tablespaces Free Space Fragmentation 330 Top Blocked Objects 334 Top SQLs 338 Top SQLs Full Text 340 Top SQLs Text 340 Usability 345 collections overview 224 components shared by all Foglight for Oracle screens 50 configuring tablespace monitoring read-only tablespaces 189 system-sysaux tablespaces 189 temporary tablespaces 189 undo tablespaces 189 connection details screen 166 current activity monitoring 97 current storage monitoring 98 D dashboard activity 122 instance locks 128 RAC locks 125 databases 47 I/O 68 instance 74 RAC 69 instance home page 57 overview 44 RAC homepage 57 storage 136 all datafiles 140 all tablespaces 138 archive logs 148 datafile-specific 144 instance redo & archive performance 90, 149 RAC redo & archive performance 88 redo & archive performance 83, 87 redo logs 146 tablespace-specific 140 toolbar 57 dashboards Oracle 44 Data Guard service monitoring 111 database discovery wizard 24 database backup panel 150 database discovery overview 10 databases dashboard 47 pluggable 135 DBO Sessions Summary collection 323 Managing Oracle Database Systems: User and Reference Guide 352 Index DB_Oracle agent properties 162 DB_Oracle agent about 161 overview 162, 164 DB_Oracle_RAC agent properties 163 DB_Oracle_RAC agent overview 162 DB_Oracle_RAC_Instance agent properties 163 DB_Oracle_RAC_Instance agent overview 163 disk I/O panel 68 disk storage monitoring 106 DSK key enabling OS monitoring 33 E email notifications See also alarms 175 configuring 175 configuring the mail server 181 defining recipients using variables 185 defining schedules 185 defnring recipients and message 183 enabling 182 enabling OS monitoring using DSA key 33 error log panel 155 Event Wait collection 251 event wait collection 251 Exadata adding agents for monitoring Exadata environments 42 cell offload efficiency 159 cell performance 159 InfiniBand topology 160 modifying the Exadata monitoring settings 44 performance drilldown 159 summary drilldown 157 Exadata cells flashcache hit ratio 159 memory utilization 159 percentage of busy CPU 159 Exadata tab 53 F Filesystem Disk collection 259 filtering display by severity 55 Flashback Area collection 260 Foglight for Oracle architecture 21 G generating reports 101 H health check report 200 historical activity monitoring 99 History Locks Tree Instance collection 261 History Locks Tree RAC collection 263 home page toolbar 57 host activity monitoring 112 Host Properties collection 265 I I/O activity report 200 I/O dashboard 68 I/O instance dashboard 74 I/O RAC dashboard 69 Initialization Parameters collection 266 Instance General collection 268 instance I/O dashboard 74 instance identification 114 instance usability panel 123 Invalid Objects collection 268 Invalid Objects List collection 270 invalid objects panel 149 investigating performance using the node tree 59 IO Activity collection 270 J Jobs Queue collection 277 L Latches Activity collection 278 locks (RAC) dashboard 125, 128 Locks Activity collection 279 locks activity collection 279 Managing Oracle Database Systems: User and Reference Guide 353 Index Locks Breakdown collection 280 locks instance collection 281 Locks_Instance collection 281 logical reads 203 M memory activity monitoring 119 Monitoring Interconnect, cluster, and global cache 106 monitoring alarms 99 cluster’s I/O activity 107 cluster’s usability and load balance 104 CPU load 108 current activity 97 current storage 98 Data Guard service 111 Dataguard physical standby 117 disk storage 106 general sessions statistics 116 historical activity 99 host activity 112 instance’s availability 112 memory activity 119 process activity 119 RAM 108 resource utilization 108 session activity 109 workload 110, 118 monitoring the instance 116 O Oracle dashboards 44 OS General Network Summary collection 293 overview adding and configuring agents 10 auto discovery 10 collections 224 dashboard 44 DB_Oracle agent 162, 164 DB_Oracle_RAC agent 162 DB_Oracle_RAC_Instance agent 163 rules 222 P PA Instance Statistics collection 293 PA Usability collection 296 Parallel collection 297 PeopleSoft monitoring 61 Performance Analysis connection status 100 performance toolbar 62 Pluggable Database State collection 301 pluggable databases monitoring 135 Pluggable Databases drilldown 135 process activity monitoring 119 R RAC I/O dashboard 69 RAM monitoring 108 read-only tablespaces 189 Redo Instances Misc collection 303 redo instances miscellaneous collection 303 redo logs panel 146 registry variables defining email notification schedules 185 defining for email recipients 185 reports report types cluster balance 199 I/O activity 200 instance availability 200 Oracle executive summary 200 PL/SQL blocks 200 storage summary 200 top DB users 200 top SQLs 200 workload summary 200 reports dashboard 199 resource utilization monitoring 108 reviewing SGA settings at the instance level 152 at the RAC level 151 reviewing the initialization parameters in instance mode 153 in RAC mode 154 Managing Oracle Database Systems: User and Reference Guide 354 Index rules See also alarms 175 overview 222 S sensitivity level, alarms 176 session activity monitoring 109 Session Details Datafile I/O PA collection 305 Session Details Locks PA collection 306 Session Details Open Cursors PA collection 307 Session Details SQL Summary PA collection 308 Session Details Statistics PA collection 309 Session List PA collection 312 sessions pane 116 sessions summary collection 323 severity filtering display by 55 severity levels editing for alarms 179 storage all datafiles dashboard 140 all tablespaces dashboard 138 archive logs dashboard 148 dashboard 136 datafile-specific dashboard 144 redo & archive performance instance dashboard 90, 149 RAC dashboard 88 redo & archive performance dashboard 83, 87 redo logs dashboard 146 tablespace-specific dashboard 140 storage summary report 200 sudo preparing a sudo environment for UNIX and Linux users 25 support 8 system-sysaux tablespaces 189 T tablespaces and datafiles panel 137 Tablespaces Free Space Fragmentation collection 330 technical support 8 temporary tablespaces 189 thresholds editing for alarms 179 toolbar dashboard 57 Top Blocked Objects collection 334 Top SQLs collection 338 Top SQLs Full Text collection 340 top SQLs report 200 Top SQLs Text collection 340 tracking changes with selected categories 61 tracking physical I/O activity 121 tracking throughput 98 U undo tablespaces 189 Usability collection 345 user-defined collections panel 134 W workload monitoring 110, 118 workload summary report 200