Transcript
easyRAID Generic Software Manual
SCSI(Ultra 320 ) / FC(2Gb) to Serial ATA II
Disk Array Systems August 2005 Version 1.0
Ultra 320 SCSI to Serial ATA II Disk Array System
& Fibre Channel to Serial ATA II Disk Array System
easyRAID Generic Software Manual
Table of Content Chapter 1
1-1
1.1 About this manual 1.2 RAID terminology
1-1 1-1
Chapter 2
2-1
2.1 System diagnostic during boot cycle
2-1
2.2 LCD screen symbols (drive status) 2.3 Menu navigation and mute system beeper from front panel
2-2 2-3
2.4 Quick Setup – Flow chart and procedures
2-4
Chapter 3
3-1
3.1 Advanced Setup - Flow chart and procedures
3-1
3.1.1 Setup the connection to RAID system 3.1.2 Setting real time clock(RTC) and checking drive health
3-2 3-6
3.1.3 Creating and slicing arrays 3.1.4 Assign SCSI/FC Channel ID & Mapping LUN to hosts
3-7 3-9
3.1.5 Save Configuration & System initialization 3.1.6 Connecting the RAID system to a host computer
3-11 3-11
3.2 Deleting or reconfiguring an array 3.3 Deleting a slice of an array
3-12 3-12
3.4 Expanding an Array 3.5 Regenerating an arrays parity and performing a parity check
3-13 3-13
3.6 Configuring SAN masking for FC disk array
3-13
Chapter 4
4-1
4.1 Disk Scrubbing – Bad block detection and parity correction 4.2 Auto Shutdown – RAID system in critical condition
4-1 4-4
4.3 Disk Self Test – Check drive healthy in RAID system 4.4 Disk Clone – Manually cloning a failing drive
4-5 4-7
4.5 SMART – Predicting a drive failure and automatic disk cloning 4.6 AV Streaming - Performance-critical streaming application
4-8 4-10
4.7 PreRead cache - Enhancing sequential read throughput 4.8 Alignment Offset – Multipath IO S/W (PathGuard) on Window OS
4-11 4-12
Chapter 5
5-1
5.1 Event Severity 5.2 Event List
5-1 5-1
Appendix
I
Upgrading Firmware of the RAID System
I
Pre-configured RAID parameters On-line and Off-line effective RAID parameters Recording of RAID Configuration
V VI VII
C
i
Revision History Version 1.0
Date Aug, 2005
Remarks Initial release
ii
Chapter 1 Introduction This chapter contains an overview of easyRAID disk array generic software. It contains the following topics: z
Section 1.1, “About this manual“
z
Section 1.2, “RAID terminology”
1.1 About this manual This manual has the following chapters and appendix: Chapter 1, Introduction, describes RAID terminology and basic SCSI concepts. Chapter 2, Quick Setup, introduces the procedure to create single array using the front panel Chapter 3, Advanced Setup, introduces advanced configuration flow to create and delete multiple arrays and slices through RS-232 and RJ-45 port Chapter 4, Using advanced functions of RAID system, contains proprietary RAID functions and configuration flow and condition. Chapter 5, Event messages of the RAID System, lists the message recorded by RAID system. Appendix, Upgrade firmware, pre-configured, online and offline effective RAID parameters.
1.2 RAID terminology Redundant Array of Independent Disk (RAID) is a storage technology used to combine multiple inexpensive drives into a logical drive to obtain performance, capacity and reliability over single-disk storage. RAID 0 - Disk Striping In RAID 0, data is divided into pieces and written to all disks in parallel. This process is called striping because the pieces of data form a stripe across multiple disks. This improves access rate, but makes availability lower, since a single disk failure will cause the failure of the array. A RAID 0 array is unsuitable for data that can not easily be reproduced, or for mission-critical system operation.
Figure 1- RAID 0(Disk striping) 1-1
RAID 1- Disk Mirroring In RAID 1, data is duplicated on two or more disks to provide high access rate and very high data availability. This process is called mirroring. If a disk fails, the RAID controller directs all requests to the surviving members.
Figure 2- RAID 1(Disk mirroring) RAID 3- Disk Striping with dedicated parity In RAID 3, data is divided into pieces and a single parity is calculated. The pieces and parity are written to separate disks in parallel. The parity is written to a single dedicated disk. This process is called striping with dedicated parity. The parity disk stores redundant information about the data on other disks. If a single disk fails, then the data can be regenerated from other data and parity disks.
Figure 3- RAID 3(Disk striping with dedicated parity)
1-2
RAID 5- Disk Striping with distributed parity In RAID 5, data is divided into pieces and a single parity is calculated. The pieces and parity are written to separate disk in parallel. The parity is written to a different disk in each stripe. Parity provides redundant information about the data on other disks. If a single disk fails, then the data can be regenerated from other data and parity disks.
Figure 4- RAID 5(Disk striping with distributed parity) RAID 6- Disk Striping with two sets of distributed parities In RAID 6, data is divided into pieces and two sets of parities are calculated. The pieces and parities are written to separate disks in parallel. The two parities are written to different disks in each stripe. Parity provides redundant information about the data on the RAID member disks. If two disks fail at the same time, the data can still be regenerated from other data and parity disks. The RAID 6 algorithm uses two independent equations to compute two sets of parity, which enable reconstruction of data when two disks and/or blocks fail at the same time. It greatly improves the data availability.
Figure 5- RAID 6 (Disk striping with two sets of parity- P and Q)
1-3
RAID TP(Triple Parity) - Disk Striping with triple distributed parity In RAID TP, data is divided into pieces and three sets of parities are calculated. The data pieces and the parities are written to separate disks in parallel. The three parities are written to different disks in each stripe. Parity provides redundant information about the data on the RAID member disks. If three disks fail at the same time, the data can still be regenerated from other data and parity disks. The RAID TP algorithms use three independent equations to compute triple parity, which enable reconstruction of data when three disks and/or blocks fail at the same time. It greatly improves the data availability.
Figure 6- RAID TP (Disk striping with triple parity – P, Q and R) JBOD – Just a Bunch of Disks JBOD stands for just a bunch of disks. In JBOD mode, the host will see each drive as an independent logical disk. There is no fault-tolerance in JBOD.
Figure 7- JBOD (Just a Bunch Of Disks)
1-4
NRAID – Non-RAID In NRAID mode, all drives are configured as a single logical drive without fault-tolerance. The total capacity of NRAID will be the sum of each drives.
Figure 8- NRAID (Non-RAID) RAID 0+1 – Disk striping with mirroring RAID 0+1 is a combination of RAID 0 and RAID 1 to form an array. RAID 30 – Striping of RAID 3 RAID 30 is a combination of RAID 0 and RAID 3 to form an array. It provides better data redundancy compared with RAID 3. RAID 50 – Striping of RAID 5 RAID 50 is a combination of RAID 0 and RAID 5 to form an array. It provides better data redundancy compared with RAID 5.
Summary of RAID Levels The following table provides a brief overview of RAID levels. A high data availability number indicates high fault tolerance.
RAID Level
Description
RAID 0
Disk striping
RAID 1
Capacity
Data Availability
Minimum drives
N
0
1
Disk mirroring
N/N
5
2
RAID 3
Striping with dedicated parity
N-1
1
3
RAID 5
Striping with distributed parity
N-1
1
3
RAID 6
Striping with 2 sets of parity
N-2
3
4
RAID TP
Striping with triple parity
N-3
4
4
RAID 0+1
Disk striping with mirroring
N/2
2
4
RAID 30
Striping of RAID 3
N-2
2
6
RAID 50
Striping of RAID 5
N-2
2
6
NRAID
Non-RAID
N
0
1
JBOD
Just a bunch of disks
N
0
1
1-5
SCSI specification The RAID system supports standard SCSI specification as below. Note that it’s recommended to use an external SCSI cable as short as possible. SCSI Type
Data bit
Data Rate
Max Cable Length
Max SCSI devices
SCSI-1
8
5 MB/s
6m
7
Fast SCSI
8
10 MB/s
3m
7
Fast Wide SCSI
16
20 MB/s
3m
15
Ultra SCSI
8
20 MB/s
1.5 m
7
Ultra Wide SCSI
16
40 MB/s
1.5 m
15
Ultra 2 SCSI
8
40 MB/s
12 m
7
Ultra 2 Wide SCSI
16
80 MB/s
12 m
15
Ultra 160
16
160 MB/s
12 m
15
Ultra 320
16
320 MB/s
12 m
15
1-6
Chapter 2 Quick Setup This chapter introduces a simple way to configure the RAID system. It is written for those who use easyRAID RAID system for the first time and want to create an array quickly. It includes these topics: z Section 2.1, “ System diagnostic during boot cycle ” z Section 2.2, “ LCD screen symbols (drive status)” z Section 2.3, “ Menu navigation and mute system beeper from front panel ” z Section 2.4, “ Quick Setup - Flow chart and procedures ”
2.1 System diagnostic during boot cycle Figure 9 displays a flow chart of the RAID system self-test on boot-up. During boot-up, the RAID system executes CPU, peripheral device, host and, disk chipset initialization. It consists of three modes (A, B, and C). A - Initial RAID status The RAID system has not been configured. Drives have not been installed; there is no RAID configuration present in the hard drives and RAID controllers NVRAM. B - RAID system initializing The RAID system is configured and drive initialization has begun. C - RAID system exists The RAID system has been configured. There is a RAID configuration present in hard drives and RAID controllers NVRAM.
2-1
Figure 9 - System Diagnostic Flow
Note: The RAID configuration is stored both in a reserved area of the hard drives and RAID controller’s NVRAM after an array set is created.
2.2 LCD screen symbols (drive status) The LCD screen shows an overview of RAID system drives status. This section explains the meaning of symbols that appear on the LCD screen. Symbol R I S X W
A C 1~8 J
Meaning The drive has an error or a fault RAID system is identifying the drive Global hot-spare drive No drive installed or drive is offline Warning – Too many bad sectors or unrecoverable data blocks in the drive or drive triggered “SMART failure warning” after running Disk SMART test The drive is being added to an array either during online expansion or rebuilding A clone drive (target drive) Array group number that a drive belongs to The drive is in JBOD mode ( No configuration mode) 2-2
Below are some examples. Example z z
RAID initial status
z
Two arrays with one global hot-spare drive
z z z z z z
Array1@RAID Level 5
z
z z
Array2 not created yet
z
Description Model Name: easyRAID Q16 Disk 1 to 16 are all in JBOD mode ( No RAID configuration mode)
LCD Screen
Model Name: easyRAID Q16 Disk 1 to 9 are members of Array1 Disk 10 to 14 are members of Array2 Disk 15 is a global hot-spare drive Disk 16 is not installed or offline Array 1 RAID Level 5 Disk 1 to 9 are members of Array1
Array 2 RAID Level X not available No drive member exists
2.3 Menu navigation and mute system beeper from front panel The RAID system can be configured using the front panel function keys.
Menu Navigation from LCD Panel Key
Description “UP” arrow key. To select options or scroll forward each character during data entry “Down” arrow key. To select options or scroll back each character during data entry
ESC
To escape or go back to previous menu screen
Enter
To go to a submenu or to execute a selected option
Mute System beeper from LCD panel The RAID system emits a beeping sound periodically when an error or failure occurs in the disk array. This audible alert can be turned off temporarily for the current event by pressing “UP” and “Down” key twice simultaneously. The beeper will activate again if a new event occurs.
2-3
2.4 Quick Setup – Flow chart and procedures This section provides a quick way to configure the RAID system with one-click using the front panel. The RAID system will automatically create a single array and map it to its first host channel. Before you begin It’s not recommended to connect the RAID system to the host computer before completing configuration and initialization. If connected during configuration process, resetting the controller could lead to occasional host server error messages like parity error or synchronous error.
Note: To ensure power redundancy, connect the power supplies to a separate circuit, i.e. one to a commercial circuit and one to a UPS (Uninterruptible Power Supply) Using the unit without a UPS will greatly increase the chances of data and RAID configuration corruption.
Quick Setup Flow Figure 10 shows the quick setup from the front panel function keys.
Figure 10- Quick Setup Flow
2-4
Quick Setup Procedures Step1. Insert drives into RAID system Make sure all of the drives are mounted securely to the disk trays and insert disk trays into disk slots Step2. Power ON RAID system Note: After powering on. the LCD status should display as seen to the right. Ex: Model Name: easyRAID Q16 Step3. Enter password “0000” to login Main Menu from LCD Press “Enter” key. “Enter Passwd” message will appear on the LCD screen. Use “ ”or“ ” to select each character and select “0000”, press “Enter” to proceed Note: “0000” is the default password. Step4. Select RAID level in Quick Setup Go to “Quick Setup->RAID Level” Use “ ”or “ ” key to select the desired level. For example: 6+spare, Select “Yes” and press ”Enter” to proceed. Note: If a spare drive is not selected or reserved, all of the installed disks will be configured as a drive member of Array 1. RAID level options in quick setup: 0/1/3/3+spare/5/5+spare/6/6+spare /TP/TP+spare/0+1/30/50/NRAID TP: Triple Parity allows three drives to fail in a single array Step5. Automatic system initialization After setup, the system will begin to initialize. It may take several hours depending on the total capacity of the array. Step6. Connecting host computer to 1st host channel of RAID system Power off RAID system and connect 1st host channel of RAID system to host computer. Power on RAID, wait for the unit to completely power on, then power on host computer.
Note: Before connecting RAID system(SCSI) to host server(step 6) 1. Check the SCSI/FC ID of RAID system and the HBA to ensure that they are not sharing the same ID with another device on the same channel. The default SCSI ID of RAID system is 0, Fibre ID setting is Auto. 2. In a SCSI daisy chain, SCSI termination is required on the last SCSI device. Make sure it is properly terminated on SCSI bus. 3. Before you disconnect or connect a SCSI cable from the RAID system, power off the host computer, than the RAID system for safety. Also note, SCSI bus does not support hot-plug function. 2-5
Chapter 3 Advanced Setup The RAID system can be configured using a terminal emulation program, such as HyperTerminal in Windows. This allows the user to create multiple RAID arrays and slices via RS-232 or RJ-45 port. It includes the following topics: z Section 3.1, “Advanced Setup - Flow chart and procedures” z Section 3.2, “Deleting or reconfiguring an array” z Section 3.3, “Deleting a slice of an array” z Section 3.4, “Expanding an array” z Section 3.5, “Regenerating an arrays parity and performing a parity check” z Section 3.6, “Configuring SAN masking for FC disk array”
3.1 Advanced Setup - Flow chart and procedures This section introduces the advanced setup flow chart and detailed procedures to configure the RAID system for multiple array groups, slicing, and LUN mapping to host channels. There are three methods to configure the RAID system. 1. Front panel function keys (Refer to Chapter 2 – Quick Setup) 2. VT-100 emulation program via RS-232 serial port or RJ-45 LAN port 3. Cross-platform Global Net software via RJ-45 LAN port
Figure 11- Advanced Setup Flow 3-1
3.1.1 Setup the connection to RAID system In this section, it introduces how to set up a VT-100 emulation program via RS-232 and RJ-45 port. 1. Using RS-232 serial port To set up the serial port connection, follow these steps. This example will use Hyperterminal for Windows. Step1. Use a serial cable (null modem) to connect the RS-232 port of RAID system to the management console (or host computer) Step2. Start Hyper Terminal on the management console For Windows, click “Start->Programs->Accessories->Communication”, then select “Hyper Terminal” The “Hyper Terminal” window appears and if the “Location Information” window appears, skip it Click “Cancel”->”Yes”->”OK” Step3. Make a new connection When “Connection Description” dialog box appears, give the new connection a name for example: Disk Array, Click “OK” to continue
Step4. Select PC serial “COM” port Click “Connect using” field and select “COM1”, then click “OK”
Note: If you are unsure which COM port to use or if the connection does not work, repeat the above steps and try a different COM port.
3-2
Step5. Set the COM port properties In “Port Setting” tab, select Bit per second => 115200 Data bit Parity
=> 8 => None
Stop bits => 1 Flow control => None Click “OK” to continue
Step6. Display RAID utility interface Press + to display the main screen of RAID utility.
Tip: + can refresh the screen information 2. Using RJ-45 LAN port (Optional) To set up the RJ-45 LAN port connection, follow these steps. Step1. Use a RJ-45 cable to connect the RJ45 port of RAID system to an ethernet switch Step2. Start Hyper Terminal on the management console Repeat the same procedure in RS-232 setup. Step3. Make a new connection Repeat the same procedure in RS-232 setup. 3-3
Step4. Select “TCP/IP (Winsock)” connection Click “Connect using” field and select “TCP/IP (Winsock)”, then click “OK” Note: Not all of easyRAID disk array support TCP/IP (Winsock) via VT-100 emulation program, Contact easyRAID support for more information.
Step5. Assign IP address of RAID system Obtain IP address automatically The default ethernet setting of RAID system is DHCP enabled. Connect the RJ-45 port of RAID system to a dynamic (DHCP) network. An IP address will be automatically assigned to the RAID system. Use “Up” and “Down” arrow key to navigate the LCD screen to get the IP address of the RAID system. Assign IP address manually The RAID system can also be assigned with a static IP address via LCD manually. 1.
Enter password “0000” from LCD panel to login in RAID utility
2.
Use arrow key to go to “System Params->Ethernet Setup->DHCP”, select “Disable”
3.
Return to “Ethernet Setup” to set the network manually. Use “Up” and “Down” arrow key to assign, IP address: xxx.xxx.xxx.xxx Netmask: xxx.xxxx.xxx.xxx Gateway: xxx.xxx.xxx.xxx
Note: The default Ethernet setting of RAID system: IP address: 192.168.1.23 Netmask: 255.255.255.0 Gateway: 192.168.1.254 4. Key in the IP address of RAID system in “Host address” field of Hyper Terminal And set the “Port Number” to 4660. Click “OK” to continue. Step6. Display RAID utility interface Repeat the same procedure in RS-232 setup.
3-4
RAID utility Interface You will see the following RAID utility interface after you access the RAID controller via the RAID system COM port or LAN port for the first time.
Use the following hot keys to explore the menu tree or execute RAID functions. Tab A
Switch between “Main Menu” and “Output Area” window Move cursor UP
Z
Move cursor DOWN
S
Scroll UP one page in output area
X
Scroll DOWN one page in output area To escape or go back to previous menu screen
ESC Enter Ctrl+D
Enter a submenu or to execute a selected option Display or re-flash menu screen
Note: The default password to enter Main Menu of RAID system is 0000(four zeros)
3-5
Main Menu The following figure is the main menu for SCSI/FC RAID system after logging into the monitor utility using a VT-100 terminal program.
3.1.2 Setting real time clock(RTC) and checking drive health In order to perform advanced RAID functions, such as scheduled disk scrubbing, SMART disk cloning and recording the date and time of event messages, etc the real-time clock must be setup. Refer to Chapter 4 for detailed information about disk scrubbing, SMART. Real Time Clock menu tree
Step1. Go to “Main Menu->System Params->RTC” Step2. Select “Set RTC” to set the clock for RAID system in the format of “MM/DD/YY W HH:MM” Symbol
Range
Description
MM
01~12
Month
DD
01~31
Day
YY
05
Year
W
1~7
Day of the week 1: Monday; 2: Tuesday; 3:Wednesday; 4: Thursday;
HH
01~24
5: Friday; 6: Saturday; 7: Sunday Hour
MM
00~59
Minute
3-6
Checking drive health Go to “Main Menu->Utility->Disk Utility->Disk Self Test” and run “Short Self Test“ for all disks.
If the suspected disk fails to pass Short Self Test, it’s recommended to replace a new drive or run Extended Self Test to further investigate drive health. For more information about Disk Self Test, see Chapter 4, Section 4.3. 3.1.3 Creating and slicing arrays Below is the flow chart to create array groups and divide it into slices.
Figure 12- Multiple Arrays & Slices Creation flow
3-7
Array Params menu tree
The RAID system can be configured up to 8 array groups with different RAID level and each array group can be divided into a maximum of 16 slices. Follow the following steps to create and slice an array. Step1: Choose the stripe size of RAID system Go to “Main Menu->Array Params->Stripe Size” and select a stripe size based on what kind of application behavior will be applied. Recommendation: 1. Using the default stripe size, 128 sectors (64 KB), should be sufficient for most applications. 2. Choose 256 sectors (128KB) stripe size if host application is mainly sequential read/write IO 3. Choose 32 sectors (16KB) stripe size if host application is mainly random read/write IO Note: Once the stripe size is selected, it will apply to the whole RAID system. All arrays will use this specified stripe size. Step2: Decide whether or not to create an over-2TB slice (Optional). Skip this step if you do not wish to create a slice over 2 TB. Currently, there are two ways to break 2 TB limitation per slice. One is to change the sector size or enable 16 byte CDB(Command Descriptor Block) 2.1 Go to “Main Menu->Array Params->Slice Over 2 TB” and select “enable” 2.2 Select “Sector size” or “16 byte CDB” Note: 1. Currently, “Sector size” option is supported for Microsoft Window 2000/2003/2003 SP1 “16 Byte CDB” option is supported for Window 2003 server SP1 2. If Slice Over 2 TB option is disabled, after initializing an array if the capacity is over 2 TB,
3-8
the RAID system will automatically create multiple slices limited at 2TB. Step3: Create multiple arrays groups 3.1 Go to “Main Menu->Array Params“, then select an array number 3.2 Go to “RAID Level”, and select a RAID level 3.3 Select drive members of the current array. Note: Hot-spare disk is not a member of an array, reserving at least one spare drive in a unit is recommended. 3.4 Press until the main menu appears 3.5 Repeat the step 3.1~3.4 to create more array groups. Step4: Creating slices of an individual array 4.1 Go to “Main Menu->Array Params“, then select the array you want to divide into slices. 4.2 Go to “Slice”, and select the first slice - Slice 00, type the size in megabyte(MB), then press . Note: The output area will display the slice size that has been created. 4.3 Repeat Step 4.2 to create the next Slice 01 until the array is divided as planned Note: It is not allowed to create slices randomly. Create slices in ascending sequence, for example: Slice 00->Slice 01…->Slice 15 Step5: Setting initialization mode Initialization mode has two options, foreground and background. With foreground initialization, an array will be accessible after initialization is completed. With background initialization, an array will be accessible during initialization. 5.1 Go to “Main Menu->Array Params”, and select the array group 5.2 Go to “Initialization Mode”, and select a mode you prefer, for example: background, the array can be accessed during initialization. Note: It may take several hours to complete foreground initialization depending on the total capacity of an array.It’s recommended to set “foreground” mode to double check drive health during RAID system initialization. 3.1.4 Assign SCSI/FC Channel ID & Mapping LUN to hosts Every device requires a unique SCSI or Fibre channel ID. A SCSI chain can support up to 15 SCSI devices with “Wide” function enabled. A Fibre loop can support up to 125 FC devices. If there are multiple host computers that require access to the same storage device (LUN) than there must be clustering or multipath I/O software installed on these computers.
Note: easyRAID RAID also supports host-based multipath I/O software PathGuard.
3-9
SCSI/Fibre Params menu tree
Assigning SCSI or Fibre Channel ID Setting SCSI ID 1.
Go to “Main Menu->SCSI Params”, then select a SCSI channel for example: SCSI CH1
2.
Select “Set SCSI ID”, then select a SCSI ID
3.
Repeat Step 1~2 to set another SCSI Channel ID
Setting Fibre ID 1.
Go to “Main Menu->Fibre Params”, then select a Fibre channel for example: FC CH1
2.
Select “Set Loop ID”, then select “Auto” or key in a number (0~125) manually.
3.
Repeat Step 1~2 to set another Fibre Channel ID
Note: 1. Check the SCSI/FC ID of RAID system and the HBA to ensure that they are not sharing the same ID on the same channel. The default SCSI ID of RAID system is 0, Fibre ID setting is Auto. 2. In a SCSI daisy chain, SCSI termination is required at the last SCSI device. Make sure it is properly terminated on the SCSI bus. 3. QAS(Quick Arbitration Select) setting is required for Ultra 320 devices. If the QAS setting on the RAID system differs from the HBA, there will be problems accessing the RAID unit from the host. Refer to the following table to properly set QAS option in SCSI Params. The RAID system and HBA require the same QAS setting. HBA vendor
QAS setting
Adaptec/ATTO
Enable(Default)
LSI
Disable
3-10
Mapping LUN(s) to a host channel. LUN mapping is the process to make a slice visible to a desired host channels. Each LUN will appear as a storage device to the host computer. 1. Go to “Main Menu”, then select “SCSI Params” for SCSI RAID or “Fibre Params” for Fibre RAID 2. Select a specific host channel, for example: SCSI CH1 or Fibre CH1 3. Go to “Lun Map”, and select a LUN number. 4. Select an array, then select the desired slice to map to the chosen LUN Ex. SCSI Params -> SCSI CH1 -> LUN Map -> LUN 0 -> Array 1 -> Slice0 This will map Slice 0 of Array 1 to LUN 0 of SCSI Channel 1 Note: 1. The same slice may be mapped to multiple LUNs on separate host channels, but is only applicable for clustering or multipath environments. 2. Selecting “Disable” will delete a LUN map. Deleting a LUN map will not delete the data contained in the array or slice. 5. Press the key to return to the Main menu 6. Repeat step 1~5 for each slice until all slices are mapped to a LUN 3.1.5 Save Configuration & System initialization Go to “Main Menu->Save Config” and select “Save & Restart”, select “Yes” to complete RAID configuration.
Note: The RAID configuration is stored in a reserved area of the hard drives and the RAID controller’s NVRAM after the array is created. After the RAID system reboots, it will enter system initialization. It may take several hours to complete depending on the total capacity of an array. 3.1.6 Connecting the RAID system to a host computer Power off the RAID system and connect to a host computer or FC switch Power on the RAID system, after it has completely powered on, power on the host computer Note: The host computer should be the last device to power on. Summary of Advanced Setup Step 1 Step 2
Insert drives into RAID system Power ON RAID system
Step 3
Setup the connection to RAID system (via RS-232 or RJ-45)
Step 4
& Setting real time clock Checking drive health(Short Self Test)
Step 5 Step 6 Step 7 Step 8
Creating and slicing arrays Assign SCSI or FC channel ID and mapping LUNs to a host channel Save configuration & system initialization Connecting the RAID system to a host computer 3-11
3.2 Deleting or reconfiguring an array Before deleting or reconfiguring an array backup all required data. To reconfigure an array with different drive members, RAID level, or stripe size, delete the existing array and then reconfigure the array. To delete an array, follow these steps: Step1. Go to “Main Menu->Array Params”, and select the array you want to delete Step2. Go to “RAID Level”, and select “None” Step3. Select “Yes” to proceed Step4. Go to “Main Menu->Save Config” and select “Save to NVRAM”. The RAID system will not automatically reboot, and the array will be deleted immediately.
Caution: Deleting an array will destroy all data contained in that array.
3.3 Deleting a slice of an array Follow these steps to delete a slice Step1. Go to “Main Menu->Array Params”, and select the desired array Step2. Go to “Slice”, and select the last slice in the array, for example: Slice 02 Step3. Type “ 0 ” (MB) to delete Slice 02 Step4. Return to “Slice” then select “Slice 01”, type “ 0 ” (MB) to delete Slice 01 Step5. Go to “Main Menu->Save Config”, then select “Save to NVRAM” to take effect the change. Delete slices in a descending sequence. Similarly, you can follow the same steps to change the slice size. Note: The use of a third party storage resource management (SRM) software or an OS file management program to divide or stripe a slice may lead to data fragmentation that causes decreased I/O performance of the RAID system.
Caution: Deleting a slice will lose all data contained in that slice. Backup your data before deleting or changing the slice size
3-12
3.4 Expanding an Array Follow these steps to expand an array. Step1. Go to “Main Menu->Array Params->Expand Array”, and select the desired array Step2. The disk numbers will then prompt, select a number that will add to the array RAID system then enters on-line expanding process automatically After array expansion is complete, a new slice will be created from the added capacity. Refer to Section 3.1.4 to map the new slice to the desired host channel/LUN.
Note: Expanding an array will impact system performance, performing array expansion during off-peak time is recommended.
3.5 Regenerating an arrays parity and performing a parity check RAID parity might become inconsistent with data after extended periods of time. Users can re-generate array parity or perform a parity check to ensure data integrity. Follow the following steps to expand an array. Step1. Go to “Main Menu->System Params“ then select “Init Parity” to regenerate RAID parity or “Parity Check” to verify the parity consistency. Step2. Select the desired array, then select “Start” to proceed Note: After the RAID system starts parity check if a parity inconsistency is detected, the parity check will stop and report a discrepancy. Refer to Chapter 4, section 4.1 about disk scrubbing to correct parity errors.
Note: Init Parity and Parity Check can only be performed when RAID system is in an optimal condition.
3.6 Configuring SAN masking for FC disk array SAN masking is a RAID system-centric enforced method of masking multiple LUNs behind a Fibre channel port. As Figure 13 shows, with SAN masking, a single large RAID device can be subdivided to serve or block a number of different hosts that are attached to the RAID through the SAN fabric. The host servers that access a certain LUN through a particular port can be masked from accessing other LUNs through the same port. SAN masking can be setup on the RAID system or the host computers HBA. Masking a LUN at the device level is more secure than masking at the host computer, but not all RAID systems have LUN masking capability; therefore, some HBA vendors allow persistent binding at the driver-level to mask LUNs.
3-13
Figure 13- SAN Mask example SAN Mask menu tree
Follow the following steps to enable SAN mask. Step1. Go to “Main Menu->Fibre Params->SAN Mask->Supporting”, and select the desired Fibre channel, for example: FC CH1, then select “Disable“ Note: The default setting, Enable, allows all SAN host computers to access all LUNs via the fibre channel port. Step2. Go to “SAN Mapping->FC CH1->Host 1”, select “LUN0”, and select “Yes”. Step3. Repeat Step 2 to map, “Host 2->LUN1”, “Host 3->LUN2”, “Host 4->LUN3”, “Host 5->LUN4” Step4. Go to “Edit WWN Tbl”, then type 8 byte WWN of FC HBA which is installed in each host server. Refer to your FC HBA documentation for more details.
3-14
Chapter 4 Using advanced functions of RAID system This chapter further introduces the advanced RAID functions. It covers the following topics: z Section 4.1, “Disk Scrubbing – Bad block detection and parity correction” z Section 4.2, “Auto shutdown – RAID system in critical condition” z Section 4.3, “Disk Self Test – Drive health test in RAID system” z Section 4.4, “Disk Clone – Manually cloning a failing drive” z Section 4.5, “SMART- Predicting a drive failure and automatic disk cloning” z Section 4.6, “AV Streaming – Performance-critical streaming application” z Section 4.7, “Pre-Read cache – Enhance sequential read throughput” z Section 4.8, “Alignment Offset – MultiPath I/O Software (PathGuard) for Windows OS”
4.1 Disk Scrubbing – Bad block detection and parity correction Objective With the increasing capacity size of hard drives, storage subsystem vendors face the challenge of handling bad blocks and parity errors. Bad sectors can form on HDD areas that are not accessed for long periods of time. These problems may lead to unrecoverable data loss. In order to effectively solve the problem and improve data availability disk scrubbing (DS) was developed. DS can scan for bad sectors and/or parity errors in a RAID array. The RAID system reconstructs bad sectors from other sectors and re-assigns it to an undamaged area. At the same time it also detects parity inconsistency; users can decide whether or not to overwrite inconsistent parity. DS is a proactive approach to address data integrity, it maintains the RAID system in a good condition. Unrecoverable data loss As Figure 14 shows, although all of disks, disk #1~4, are online at t=t0, block number D3 is already a bad sector. Even after the rebuild process is completed at t=t3. Data block D1 will not be able to successfully regenerate from the other data and parity blocks.
Figure 14- Unrecoverable data
4-1
Parity inconsistency Over long periods of time, parity blocks may not be consistent with data blocks. This may result from unexpected power outages or resetting the RAID system before cached data is written to drives. If the parity inconsistency is detected, it indicates a data error exists either on the data disk or parity disk. However, the RAID system can not determine if the error resides on data or parity disks because of the RAID algorithm. Enabling “Overwrite Parity” in disk scrubbing will automatically correct data on the parity disk whenever parity inconsistency is detected. If the array’s parity is seriously damaged, with overwrite parity enabled, data loss may occur after disk scrubbing is completed. Disable it if the parity data has been seriously damaged. Figure 15 describes the detailed flow of disk scrubbing (DS). When DS is running, the controller will read data and parity blocks of a stripe in an array and execute a parity check. DS predicts data block failure and corrects parity errors in order to enhance data integrity and availability.
Figure 15 function description of disk scrubbing
Note: 1. Disk scrubbing can only be activated when the array is in optimal condition. This means there are no drive member failures in the array and no background task in progress, i.e. array expansion etc. 2. Disk Scrubbing will impact I/O performance; running DS during off peak times is recommended. 3. If the RAID system is powered off while DS is running, it will not resume on the next power up.
4-2
Follow the following steps to configure disk scrubbing.
Figure 16 Disk Scrubbing Configuration Flow Disk Scrubbing Menu Tree
Enabling Overwrite parity option is recommended. Overwrite parity option applies to Disk Scrubbing in both manual and scheduled mode.
Manual Scrubbing Follow these steps to manually start and stop disk scrubbing, Step1: Go to “Main Menu->Utility->System Utility->Disk Scrubbing” and select Manual Scrubbing Step2: Select All arrays or a single array Step3: Select Start or Stop. Once started, the percentage of progress is indicated on the LCD screen. Step4: Repeat the above steps to start or stop scrubbing for other arrays groups.
4-3
Scheduled Scrubbing Follow these steps to schedule disk scrubbing. Note: Enable RAID system clock in advance before setting up the schedule scrubbing, or scrubbing will not be able to activate. Refer to section 3.1.2 to setup system real time clock (RTC). Step1: Go to “Main Menu->Utility->System Utility->Disk Scrubbing” and select Schedule Scrubbing Step2: Select all arrays or a single array Step3: Select Schedule ON Step4: Select the preferred cycle to run disk scrubbing periodically. For example: Once per 4 weeks Step5: Select the Day of the week, for example: Sat Step6: Manually key in the hour of 24 hour clock, for example: 00 hour Step7: Repeat step1~6 to set up scrubbing schedule for other arrays groups.
Disk Scrubbing Report After disk scrubbing is completed, in the output window of hyper terminal, the following information will be displayed. For example: -----------------------------------------------------------------------------------------------------------Disk Scrubbing Result: --1. Bad Block Check-Disk # 1: Found 3 Bad Blocks, Recovered 2, Total 10+(3) Bad Blocks Disk # 2: Found 6 Bad Blocks, Recovered 6, Total 0+(6) Bad Blocks Disk # 3: Found 1 Bad Block, Recovered 1, Total 12+(1) Bad Blocks …………………………………………………………………. Disk # 16:Found 0 Bad Blocks, Recovered 0, Total 19+(0) Bad Blocks --2. Array Parity Check (Overwrite Parity YES) -Array X: Found 3 Parity Errors, Overwrite Parity Or --2. Array Parity Check (Overwrite Parity NO) -Array X: Found 3 Parity Errors, Overwrite Parity-NONE ----------------------------------------------------------------------------------------------------------
Description of Disk Message: Disk # a : Found b Bad Blocks, Recovered c, Total x + (b) Bad Blocks a is the disk number, b is the number of bad blocks found during this session of scrubbing, c is the number of bad blocks recovered x is the total number of bad blocks.
4.2 Auto Shutdown – RAID system in critical condition Objective The RAID system will be protected against internal overheating condition or UPS AC power
4-4
loss. In the RAID system, there are several thermal sensors located on the RAID controller and midplane. If the temperature increased to a dangerous level in the RAID system, it may damage internal components including disk drives. With auto shutdown enabled, if overheating occurred or AC power loss is detected, the RAID system will shutdown after a given time duration.
Under the following conditions, auto shutdown will activate automatically if it is enabled. 1.
System temperature exceeds threshold
2.
All fans failure or not available
3.
UPS AC Power Loss
Auto shutdown menu tree
Note: System real time clock (RTC) must be activated in order for the auto shutdown feature function properly.
Follow the steps to set up auto shutdown, Step1: Go to “Main Menu->Utility->System Utility->Auto Shutdown->Enable“ Step2: Select “Event Trigger” and enable or disable each trigger option. Step3: Select “Shutdown duration”, for example: 15 min, RAID system will turn off the power automatically after 15 minutes if shutdown event has triggered.
Note: 1. If the event triggers are disabled or critical events return to the normal, auto shutdown will be inactivated or canceled. 2. If an auto shutdown event is triggered, the write cache will change from “Auto” to “Write-through” mode to ensure data integrity.
4.3 Disk Self Test – Check drive healthy in RAID system Objective Disk Self Test (DST) is used to test the health of disks with them installed in the RAID system. Prior to DST a user would have to remove disks individually and run a vendor proprietary disk utility in a separate host computer DST predicts the likelihood of near-term HDD degradation or fault condition. DST performs write test, servo analysis and read scan test on the disks.
4-5
Follow the following steps to check disk health condition in RAID system, 'LVN6HOI7HVW
5XQ 6KRUW'67 3DVVRU)DLO" 5HSODFHZLWK QHZGULYHV )DLO
3DVV
)DLO 5XQ ([WHQGHG'67 3DVVRU)DLO" 3DVV &UHDWH$UUD\ 6OLFH
Figure 17 Disk Self Test Configuration Flow Disk Self Test menu tree
Disk Self Test: Step1: Go to “Main Menu->Utility->Disk Utility->Disk Self Test“ Step2: Select “Short Self Test” and select “All Disks” or “Disk X” to start drive self test Step3: Select “Extended Self Test”, if there is any error occurrence after short DST (Step2) Then select “All Disks” or “Disk X” to further check the suspected drive. Swap the suspected drive if it doesn’t pass Extended Self test.
Note: 1. Running DST before creating an array is recommended. DST will not overwrite data. 2. DST can only be executed in offline mode. This means that if there is any host activity DST will be terminated and host I/O access will resume. 3. DST can also be performed thru LCD function key directly. Key in password- “1111” to start short or extended self test of all drives. 4. It may take several hours to run Extended DST depending on the drive capacity and spindle speed. 5. Most newer hard drives support DST, contact your drive vendor to see whether your drives support DST.
4-6
4.4 Disk Clone – Manually cloning a failing drive Objective Hard drives are the most likely component to fail in a RAID system, and are very difficult to predict when the failure will occur. When a failure does occur the RAID unit will have to regenerate data from the non-failed hard drives to rebuild a new drive, and during this time the RAID system will be in degraded mode. This is where Disk Cloning (DC) can aid a user. Disk Clone can copy a failing drive to a hot spare, and upon completion of cloning, the new cloned disk can take the position of the failing disk immediately or can stand-by until the cloned disk fails. Disk cloning is to help prevent a rebuild from ever occurring and having the unit in degraded mode. There are two options to clone a failing drive: “permanent clone” and “swap after clone”. In "permanent clone" mode, the clone disk (hot spare disk) will be the mirror of the source disk until the source disk has failed. The clone disk will then replace the source disk. In "swap after clone" mode, immediately after the clone process is complete the clone disk replaces the source disk and the source disk is taken offline. Disk Clone menu tree
Follow the steps to start disk cloning manually, Step1: Go to “Main Menu->Utility->Disk Utility->Disk Clone“, then select “Start Disk Clone” Step2: Select “Source Disk” which is the suspected failing drive in the array. Step3: Select “Target Disk” and select a target drive. (Clone drive) Note: Only the hot-spare drive will be displayed in target disk. Step4: Select “Start Permanent Clone” or “Start Swap After Clone” to start disk clone. After DC is complete, the target disk (clone) status will be marked with a “C” on the LCD panel. Step5: Repeat these steps to clone other drives.
Note: 1.If cloning is in progress, and the source disk fails or goes offline, the cloning disk will replace the source disk and become array’s member drive. Also the RAID system will begin rebuilding at the point where cloning stopped. 2. If cloning is in progress, and a member drive fails in an array, excluding the source disk, the cloning will stop and the RAID system will begin rebuilding. 3. Disk clone can only be performed while the array in an optimal condition.
4-7
4.5 SMART – Predicting a drive failure and automatic disk cloning Objective Disk Clone (DC) is a process of manually cloning data from a source disk to target disk. With the SMART function, the RAID system monitors drive health on preset polling intervals, if hard drive degradation is detected or the user-defined bad sector threshold is reached, the cloning function will begin immediately. SMART Event Trigger There are two SMART event triggers that will begin disk cloning: a SMART failure flag, and a user-defined bad sector threshold. The SMART failure flag is triggered by the drive, and is defined by vendor-specific attributes that may differ model to model. The user-defined bad sector threshold is a specific number of bad sectors per drive. The user must input the bad sector threshold to start disk cloning. SMART Mode There are four modes in SMART function. 1.
Disable: SMART function is inactivated.
2.
Enable ( Alert Only) : The RAID system monitors using drives SMART on preset time intervals. When a SMART failure is detected, the user will be alerted with a beeper and the drive’s status will be changed to a “W” on the front LCD which indicates a warning
3.
Enable ( Permanent Clone) : The RAID system monitors drives SMART, if a SMART failure is detected or user-defined bad sector threshold is reached, disk clone will begin. Upon completion, the clone disk (hot spare disk) will be the mirror of the source disk until the source disk has failed.
4.
Enable ( Swap After Clone) : The RAID system monitors drives SMART, if a SMART failure is detected or user-defined bad sector threshold is reached, disk clone will begin. Immediately after cloning has completed, the clone disk (hot spare disk) replaces the source disk and the source disk is taken offline.
Follow the following steps to configure SMART function,
4-8
SMART menu tree
Follow the steps to configure SMART function, Step1: Go to “Main Menu->Utility->Disk Utility->SMART“, and select “Test Disk SMART” First, check whether or not your drives support SMART. If your drives don’t pass SMART test, drive’s status will change to “W” on the LCD screen. Step2: Go to “Bad Sector” and decide the value of “Threshold for Clone” or “Threshold for Swap” For example: Threshold for Clone: 130, if a drive accumulates 130 bad sectors, disk clone will start. Threshold for Swap: 200, if a drive accumulates 200 bad sectors, the source disk will be taken offline. Step3: Go to “Disk Check Time” then select a time interval to monitor drive’s SMART and bad sector status. For example: 60 minutes Step4: Go to “SMART Mode”, then select the mode you prefers.
Note: 1. Make sure system clock is enabled before configuring SMART function. 2. If bad sector threshold is disabled for Clone and Swap, disk clone will only be activated when drives a SMART failure is detected.
4-9
4.6 AV Streaming - Performance-critical streaming application Objective Bad blocks and read or write delays of drives are unavoidable in a RAID system. For AV streaming applications, such as broadcast, post production, video/audio editing application, etc., these errors will cause choppy audio and/or video frame loss. In some instances the entire RAID system will stop operation. Enabling AV Streaming option in the RAID system can eliminate the chance of data transfer delays in a performance-critical streaming application. AV stream option will shorten drive I/O response time, re-arrange cache buffer management for read/write commands and changes the algorithm to read/write data. Only enable the AV Streaming option after it has been tested in a real AV streaming environment by an experienced engineer. AV Streaming menu tree
Follow the steps to configure AV Application, Step1: Go to “Main Menu->Utility->System Utility->AV Streaming“, then select “Enable” Step2: Go to “Disk Timeout” and then select a disk I/O timeout value, for example: 3 seconds Note: Once AV streaming option has been enabled and disk timeout has been changed to a low value, the RAID system will frequently alert remapped blocks in the status log. Step3: Go to “Remap Threshold” and then select a threshold that the RAID system will start to alert remap warning message.
Note: 1. AV streaming can only be enabled for a single array configuration. If multiple arrays are present in the RAID system, AV streaming will not work. 2. Single array and single slice is the optimal RAID configuration for AV streaming. Partitioning an array into multiple slices for AV streaming is not recommended.
4-10
4.7 PreRead cache - Enhancing sequential read throughput Objective PreRead cache is used to accelerate the performance of applications that access data sequentially, such as film, video, medical imaging and graphic production industries. With PreRead cache enabled, the RAID controller move to cache the next blocks of data that will be needed in the sequence. It reads the data from slow, nonvolatile storage and places it in fast cache memory before it is requested. Only enable PreRead cache after it has been tested in a read-intensive application environment by experienced engineer.
PreRead menu tree
Follow the steps to configure PreRead function, Step1: Go to “Main Menu->Array Params->PreRead Setup”, then select “Enable” Step2: Select “Max ReadLog” and key in a number, for example: 32 Max Readlog is the record of Read commands that were issued from a host application. Step3: Select “Max PreRead” then key in a number for example: 16 Max Preread is the data depth that will be read ahead in advance.
Note: If PreRead is not properly setup, it will decrease the I/O performance of the RAID system.
4-11
4.8 Alignment Offset – Multipath IO S/W (PathGuard) on Window OS Objective Operating systems (OS) reserves private information, known as a signature block, at the beginning of a LUN. The result is an un-alignment of disk striping. As Figure-18 shows, after a physical device is formatted with a file system, a data segment may cross two stripes causing a split I/O command to complete a read or write. Alignment offset is used to set the host logical block address (LBA) alignment with the stripe boundary of a LUN so that it enhances I/O performance of the RAID system. In order to fix this problem and enhance system performance under different operating systems, the LUN should be offset based on its file system type. Setting alignment offset is recommended when using PathGuard, easyRAID’s Multipath I/O software, for Windows operating system.
Figure 18 Alignment Offset Alignment offset menu tree
Follow the steps to configure alignment offset, Step1: Go to “Main Menu->Array Params->Array x->Alignment Offset”, then select a slice Step2: Select “NTFS” if the slice will be formatted with Window NT file system
Note: 1. Currently, only Microsoft Window NT file system (NTFS) supports the alignment offset function. 2. Alignment offset should be configured before the slice is created. 3. PathGuard is a host-based multipath I/O software for Window 2000/2003 server. Contact technical support team for more information.
4-12
Chapter 5 Event Message of RAID system This chapter lists the event message recorded by the RAID system. It contains the following topics: z
Section 5.1, “Event Severity”
z
Section 5.2, “Event List”
5.1 Event Severity Events are classified with different severity levels. 1. 2. 3.
Error - Event messages that indicates a significant problem, such as drive/fan/power failure...etc Warning – Event messages that are not necessarily significant, but might indicate a possible future problem Information – Event messages that describes a successful operation of RAID function.
5.2 Event List 1.
Event severity: Error
DISK X initial fail, status 0xY !ʳʳ
Disk X initialization failed with status Y ʳ
DRAM TEST FAILʳʳ
DRAM diagnostic test failedʳ
Disk X retry SPIN_TIMEOUTʳʳ
Failed to retry operation to disk X within SPIN_TIMEOUT. The disk was offlined.ʳ
Disk X initial SPIN_TIMEOUTʳʳ
Disk X could not be ready within SPIN_TIMEOUT. The disk was offlined.ʳ
ERROR: Disk X Identify Data Error!ʳʳ
Failed to identify disk X. The disk was offlined.ʳ
ERROR: Disk X Inquiry data ERROR !ʳʳ
Invalid inquiry data on disk X. The disk was offlined.ʳ
ERROR: Disk parameters ERROR !ʳʳ
Invalid Cylinder/Head/Sector disk parameters found. The disk was offlined.
ERROR: No multi-sector mode !ʳʳ
Disk did not support multi-sector mode. The disk was offlined.ʳ One or more bits error were detected by ECC memory. The faulty address
ECC Error Detected at Address 0xXʳʳ
is at X. If there are more than one bit errors, the system hangs and the LCD shows "ECC MultiBit Err".ʳ
Error: spin IOC_READY timeoutʳʳ
SCSI chip initialization failed. Controller fault.ʳ
Error: No FreeChain. MX OYʳʳ
Magic: running out of scatter-gather resource on SCSIʳ
Host Channel X Init Fail! ʳ
Host channel X initialization failed. Controller fault.ʳ
Issue IOC Fact failed!ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
Issue IOC Init failed!ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
INIT: EnablePost X failedʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
IOC reset failedʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
IOC handle ready state failedʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
INIT: CmdBufferPost failedʳʳ
SCSI chip initialization step failed. Controller fault.ʳ 5-1
INIT: EnableEvents failedʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
INIT: EnablePost X failedʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
LQ_CRC_ERR..ʳʳ
SCSI chip indicated CRC error during LQ-nexus IU transfer. ʳ
MSG_OUT_PARITY_ERR..ʳʳ
SCSI chip indicated parity error during message-out phase transfer. ʳ
Member Disk#X is Warning!ʳʳ
The disk remapped entries of the Array member X has reached the threshold. The scrubbing procedure is canceled.ʳ
NVRAM :0xX Error!ʳʳ
NVRAM testing failure at address Xʳ
NVRAM TEST FAILʳʳ
NVRAM diagnostic test failed. Controller fault.ʳ
Overwrite Fail: DevID=0xX, BlkNo=0xYʳʳ
Fail to overwrite parity while scurbbing a RAID6/TP Arrayʳ
Pci SErr Assert[0xX]: 0xYʳʳ
PCI bus error code reported by SATA chipʳ
Pci SErr Cause [0xX]: 0xYʳʳ
PCI bus error code reported by SATA chipʳ
Param checksum ERROR!ʳʳ PCI BUS Error Total count:Xʳʳ PROTOCOL_ERR..ʳʳ
NVRAM superdata checksum error.
This can be resulted from firware
upgrade or NVRAM malfunctioned.ʳ Accumulation count of PCI bus error reported by SATA chip is Xʳ SCSI chip indicated protocol error found. This often occurs due to the previous signal quality issues.ʳ
Parity Overwrite
Failed to overwrite parity while scurbbing a
Fail:DevID=0xX,BlkNo=0xYʳʳ
RAID5/RAID3/RAID50/RAID30 Arrayʳ
Power module1: Failʳʳ
Power module 1 of the chassis failedʳ
Power module2: Failʳʳ
Power module 2 of the chassis failedʳ
PortX: issue PortFact failedʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
PopulateReplyFreeFIFO failed!ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
RAID30/RAID50 Init ERROR!ʳʳ
RAID30/RAID50 Array background initialization failed.ʳ
R6 X: Error!!More than 2 errorsʳʳ
More than 2 errors were found on a RAID6 Array. Failed to rebuild the Array.ʳ
RAID3/RAID5 Init ERROR!ʳʳ
RAID3/RAID5 Array background initialization failed.ʳ
RPG ERRORʳʳ
XOR engine reported error. Controller fault.ʳ
ReceiveDataViaHandshake Fail1ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
ReceiveDataViaHandshake Fail2ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
ReceiveDataViaHandshake Fail3ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
ReceiveDataViaHandshake Fail4ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
SATA Chip X Pci Err ! main_int=0xYʳʳ
PCI bus error detected by SATA chip X with interrupt status Y.ʳ
Scrub: I/O Error, Skip Row Xʳʳ
Failed on scrubbing a RAID6/TP Array due to more than 2/3 errors found.ʳ
SATA Chip X failed: Y Zʳʳ
SATA chip X initialization failed. Controller fault.ʳ
SendMessageViaHandShake Fail1ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
SendMessageViaHandShake Fail2ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
SendMessageViaHandShake Fail3ʳʳ
SCSI chip initialization step failed. Controller fault.ʳ
Send_HandShake_Request Fail1ʳʳ
SCSI chip handshake I/O failed. Wait system retrial.ʳ
Send_HandShake_Request Fail2ʳʳ
SCSI chip handshake I/O failed. Wait system retrial.ʳ 5-2
Send_HandShake_Request Fail3ʳʳ
SCSI chip handshake I/O failed. Wait system retrial.ʳ
Send_HandShake_Request Fail4ʳʳ
SCSI chip handshake I/O failed. Wait system retrial.ʳ
TP X: Y Errors!!More than 3 errorsʳʳ
More than 3 errors were found on a TP Array. Failed to rebuild the Array.ʳ
Warning: Source disk X error!ʳʳ
Failed reading source disk X during clone-processʳ
Warning: Target disk X error!ʳʳ
Failed writing target disk X during clone-processʳ
Warning:The start sector is incorrect!ʳʳ
Magic: the cloning block address is out of rangeʳ
Warning: DiskX's remap area is full!ʳʳ
Remap area of the cloning target disk X was full. The disk was offlined.ʳ
2.
Event Severity: Warning
Array X:Found Y Parity Errors,
Y parity errors were found on Array X during scrubbing. Overwrite parity
Overwrite Parity
automatically.
Array X:Found Y Parity Errors,
Y parity errors were found on Array X during scrubbing. Skip writing parity.
Overwrite Parity-NONE Disk X ERROR: Y Block 0xZ
Disk X read/write test error at block Z using test mode Y
Disk#X SMART Enable Fail!
Failed to enable SMART function of disk X
Disk#X SMART Disable Fail!
Failed to disable SMART function of disk X
DISK#X DST Fail!
Disk X DST(Disk Self Test) failed
Disk#X: DST Completed, unknown
Disk X DST(Disk Self Test) completed with unknown error.
failure, FAIL Disk#X: DST Completed with
Disk X DST(Disk Self Test) completed with electrical failure.
Electrical failure, FAIL Disk#X: DST Completed with Servo
Disk X DST(Disk Self Test) completed with servo failure.
failure, FAIL Disk#X: DST Completed with Read
Disk X DST(Disk Self Test) completed with read failure.
failure, FAIL Disk#X: DST Completed with handling
Disk X DST(Disk Self Test) completed with handling failure.
failure, FAIL Some disk does not support 48-bit LBA and current stripe size is over 256 ERROR: Disk not support LBA48 addressing! Error occurs when zeroing disk X!
sectors. Please replace the disk or set the stripe size lower or equal to 256 sectors. Failed to zero disk X for cloning process Wrong Gateway IP address format. The legal format is xxx.xxx.xxx.xxx
Gateway IP Set Error
where xxx is a decimal value from 0 to 255
IDE_ISR_1(X): status 0xY, error: Z !!
Error found on disk X with status Y, error code Z.
IDE_ISR_2(X): status 0xY, error: Z !!
Error found on disk X with status Y, error code Z.
Input(X) error, LUN # must be 0 ~ 127.
Please number the LUN from 0 to 127 Wrong IP address format. The legal format is xxx.xxx.xxx.xxx where xxx is a
IP Address Set Error Input(X) error, Host # must be 1 ~ 32.
decimal value from 0 to 255 Please number the host computer from 1 to 32 5-3
Modem operation timeout. The ongoing faxing or paging operation would try Modem timeout !
again.
Param vender ID ERROR!
Vendor ID in NVRAM mismatch. It is happened on first time system startup.
Parity ERROR:blk 0xX !!
Parity error at block X when parity check is in process
Parity P check error, RowBlkNo=X
Error found on Parity P when parity check is in process
Parity Q check error, RowBlkNo=X
Error found on Parity Q when parity check is in process
Parity R check error, RowBlkNo=X
Error found on Parity R when parity check is in process
Parity ERROR Disk#X Blk: 0xY !!
RAID3/RAID5 Array srcubbing error on disk X at block Y Wrong date-time format. The legal format is xx/xx/xx x xx:xx where x is a
RTC Parameters Error!!
decimal value from 0 to 9. Wrong Month format. The legal format is mm/xx/xx x xx:xx where mm is a
RTC Parameters Month Error!!
decimal value from 1 to 12. Wrong Day format or violating the perpetual calendar. The legal format is
RTC Parameters Day Error!!
xx/dd/xx x xx:xx where dd is a decimal value from 1 to 31. Wrong Year format. The legal format is xx/xx/yy x xx:xx where yy is a
RTC Parameters Year Error!!
decimal value from 1 to 100. Wrong Day of Week format or violating the perpetual calendar. The legal
RTC day of week Error!!
format is xx/xx/xx w xx:xx where w is a decimal value from 1 to 7. They stand for Monday to Sunday by increasing order respectively. Wrong Hour format. The legal format is xx/xx/xx x hh:xx where hh is a
RTC Parameters Hour Error!!
decimal value from 1 to 24. Wrong Minute format. The legal format is xx/xx/xx x xx:mm where mm is a
RTC Parameters Minute Error!! RAID30/RAID50 check ERROR! RAID6/RAID TP CHECK ERROR,
decimal value from 1 to 60. RAID30/RAID50 Array parity check failed Parity checked error on the RAID6/TP Array X.
RAID=X RAID3/RAID5 check ERROR!
RAID3/RAID5 Array parity check failed
Subnet Mask Set Error
Wrong subnet mask format. XXX is a decimal value from 0 to 255. IDE timer expired. This reveals the accumulated timeout couts and lost
Timeout: X, Lostint: Y The IDE Timeout value must between 1
interruption count. The specified IDE command timeout value should be from 1 to 60 seconds.
and 60 seconds!! The RTC (Real Time Clock) was not started. Please start it by setting correct Warning RTC Not Working!!
time.
5-4
Appendix Upgrading Firmware of RAID System Pre-configured RAID parameters On-line/Off-line effective RAID parameters Recording of RAID Configuration
Upgrading Firmware of the RAID System The firmware of RAID system can be upgraded via LAN port. Contact Technical support team for the latest RAID firmware. Note: Group F/W upgrade will be supported soon. Step1. Use a serial cable to connect the RS-232 port of the RAID system and the management console (host computer) Repeat the same step in Section 3.1.1 – Using the RS-232 serial port to set the RS-232 serial parameters and start the Hyper Terminal program. Make sure the connection is linked and logged into the Monitor utility interface. Step2. Use a RJ-45 cable and set up a connection to the LAN port of the RAID system and the management console via an ethernet switch
Ex: IP address of console: 10.10.4.85 (FTP server) IP address of RAID system: 10.10.4.88 2.1 Connect to the RAID system from your management console via an Ethernet switch 2.2 Set up a FTP server on your console and assign the firmware file path. For example: Use a freeware FTP server, Tftp32, and select the directory where the latest firmware is located.
I
Step3. Login to the Boot Utility Menu of RAID system via terminal program Reboot the RAID system and while RAID system is self-test, press + to enter RAID’s Boot utility interface via Hyper Terminal program.
Step4.Set IP address of the console and RAID system In “Boot Utility Menu”, press to set IP address for the console and RAID system For example: Local IP address: 10.10.4.88(RAID system) Server IP address: 10.10.4.85(Console)
II
Step5. Load firmware image file to the RAID system Press and then the file name prompt. Type the firmware file name and press Enter to load image file from the console(FTP server) to RAID system cache. Note: Make sure the firmware file name and path is assigned and correct.
Step6. Update firmware to system ROM Press to update firmware to RAID controller ROM. And press to start writing new system firmware to ROM. Note: There are dual Flash ROMs, a main ROM and a backup ROM on the RAID controller. This helps the RAID controller recover from any issues that may happen during the firmware upgrade or single ROM failure.
III
Step7. Reboot RAID system Once the firmware has updated to Flash ROM 1 and ROM 2, Press to reboot the RAID system. After reboot, the RAID system firmware will be updated.
IV
Pre-configured RAID parameters The following table lists RAID parameters that should be determined at the initial stage of RAID system configuration. 1. Quick Setup Parameter RAID Level
Default Setting 5
Alternative 0/1/3/3+spare/5+spare/6/6+spare/TP/TP+spare/0+1/30/50/NRAID
2. Array Params Parameter
Default Setting
Alternative
RAID Level
5
0/1/3/3+spare/5+spare/6/6+spare/TP/TP+spare/0+1/30/50/NRAID
Slice
Slice 00 (max 2TB)
Slice over 2 TB
Initialization Mode
Foreground
Background
Stripe Size
128 sector (64KB)
8/16/32/64/256/512/1024 (sectors)
Sector Size (Slice Over 2 TB) 16 byte CDB (Slice Over 2 TB) Sector per Track
1024(4TB) Byte
2048 (8TB) Byte 4096(16TB) Byte
16 byte CDB 255
128
3. SCSI Params ( For SCSI host interface) Parameter
Default Setting
Alternative
Set SCSI ID
0
1~15
QAS
Enable
Disable
4. Fibre Params (For Fibre host interface) Parameter Set Loop ID
Default Setting Auto
Alternative Manually
5. System Params Parameter Ethernet Setup Set RTC
Default Setting DHCP enabled
Alternative Manually set IP/Netmask/Gateway
MM/DD/YY W HH:MM
V
On-line and Off-line effective RAID parameters Type
Description RAID parameters that
I
RAID Parameter 1. SCSI Params->SCSI CH1/Ch2->Set SCSI ID, Speed, Wide, LUN Map, QAS
need to save to NVRAM and reset RAID system to take effect
II
1.
Array Params->Array x->RAID Level
2.
Array Params->Array x->Slice
3.
Array Params->Write Cache
4.
Array Params->PreRead Setup->Max ReadLog, Max PreRead
5.
Array Params->Slice Over 2TB
RAID parameters that
6.
Array Params->Sector per Track
only need to save to
7.
Fibre Params->FC CH1/CH2->Set Loop ID, Connect Mode, Set Data Rate, LUN
NVRAM to take effect
Map 8.
Fibre Params->FC CH1/CH2->SAN Mask
9.
System Params->RS232 Params
10. System Params->Passwd Setup 11. System Params->Ethernet Setup->DHCP, IP Address, Netmask, Gateway 12. System Params->Beeper
III
1.
Main Menu->Quick Setup
2.
Array Params->Expand Array->Array x->Select Disk Number
3.
System Params->RTC->Set RTC
RAID parameters that
4.
System Params->Init Parity
take effect directly without
5.
System Params->Parity Check
saving to NVRAM or
6.
Utility->System Utility->Disk scrubbing
resetting RAID system
7.
Utility->Disk Utility->Disk Self Test
8.
Utility->Disk Utility->Disk Clone
9.
Utility->Disk Utility->SMART
10. Main Menu-Shutdown
VI
Recording of RAID Configuration System Information Product Model Name Firmware Version Serial Number Installed Memory (MB) Hard Drive Information ( Vendor/Model) HDD 1
HDD 9
HDD 2
HDD 10
HDD 3
HDD 11
HDD 4
HDD 12
HDD 5
HDD 13
HDD 6
HDD 14
HDD 7
HDD 15
HDD 8
HDD 16 Ethernet Information
IP Address Netmask Gateway Mac Address Array Groups Information Stripe Size (KB) Write Cache (Auto/Enable/Disable) Array Group (1~8)
RAID Level
Slice (0~15)
Capacity (GB)
VII
Hot Spare (Yes/No)
RAID Member
SCSI Channel Information SCSI Channel (CH1/CH2)
SCSI ID (0~15)
Speed (Ultra x)
Wide (Enable/Disable)
QAS (Enable/Disable)
Fibre Channel Information Fibre Channel
Loop ID
Connection Mode
Date Rate
(CH1/CH2)
(Auto/Manual)
(FC-AL/Pt-to-Pt)
(1Gb/2Gb/Auto)
SCSI/FC Channel
LUN #
Array #
Slice #
Capacity
(CH1/CH2)
(0~127)
(1~8)
(00~15)
(GB)
LUN Mapping
VIII
