Ibm System Storage N Series: Virtual Storage Console 4.2 For

Transcript

IBM System Storage N series
 Virtual Storage Console 4.2 for VMware vSphere Release Notes GC52-1351-10 Contents Virtual Storage Console 4.2 for VMware vSphere release notes . . . . . . . . . . . . . 1 New and changed features. . . . . . . . . . . . . . . . . . . . . . . . . . 5 VSC for VMware vSphere and other products . . . . . . . . . . . . . . . . . . . . 5 Virtual Storage Console for VMware vSphere limitations . . . . . . . . . . . . . . . 7 High-latency networks affect time required for rendering . . . . . . . . . . . . . . . . 7 VSC for VMware vSphere halts when installed on vCenter Server managing multiple ESX/ESXi hosts 7 Limitations that affect Monitoring and Host Configuration . . . . . . . . . . . . . . . 8 Limitations that affect Provisioning and Cloning . . . . . . . . . . . . . . . . . . . 9 Limitations that affect the Optimization and Migration capability . . . . . . . . . . . . . 10 Limitations that affect the Backup and Recovery capability and Restore Agent . . . . . . . . . 11 Known issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Known issues with VSC for VMware vSphere capabilities . . . . . . . . . . . . . . Data ONTAP vsadmin role insufficient for VSC for VMware vSphere tasks . . . . . . . . No IP address found if VMkernel and storage systems not in same network . . . . . . . VSC for VMware vSphere log export and file downloads fail when using IE 9. . . . . . . Volume container not destroyed even if the last LUN had been deleted . . . . . . . . . VSC for VMware vSphere fails to establish a secure connection with storage system. . . . . Restarting the service delays start of VSC for VMware vSphere for 90 seconds . . . . . . Application configuration error . . . . . . . . . . . . . . . . . . . . . . Out of memory or script errors . . . . . . . . . . . . . . . . . . . . . . Error message "permission denied" when accessing the vSphere Client . . . . . . . . . Clicking several links concurrently in the navigation pane results in an error . . . . . . . Error message "A script on this page is causing Internet Explorer to run slowly..." . . . . . Problem occurs when starting the vSphere Client if Virtual Storage Console server is unavailable Security certificate warnings displayed . . . . . . . . . . . . . . . . . . . . Known issues with Monitoring and Host Configuration . . . . . . . . . . . . . . . Status of directly connected Vserver incorrectly reported in Monitoring and Host Configuration. ESXi might finish alignment with different LBAs than stated with the -- preview option . . . mbralign for ESXi takes a long time to align files . . . . . . . . . . . . . . . . VM does not boot after running mbralign . . . . . . . . . . . . . . . . . . . MPIO Settings still shows Alert icon after running Set Recommend Values . . . . . . . . Windows cluster disk reservation lost after faults in ALUA configuration . . . . . . . . Windows guest OS might crash during storage faults because of virtual memory use . . . . ESX with Windows Server 2003 MSCS guest uses partner path after controller giveback . . . ESX or ESXi 4 in an FC or FCoE configuration with ALUA might show all dead paths for some LUNs after a controller fault . . . . . . . . . . . . . . . . . . . . . . . Virtual machine disk I/O throughput can drop without sufficient physical memory . . . . . Cisco 10-Gb network or FCoE switch port in error disabled state . . . . . . . . . . . vSphere Client shows inactive datastores for ESX/ESXi hosts . . . . . . . . . . . . Known issues with Provisioning and Cloning . . . . . . . . . . . . . . . . . . XenDesktop export file placed in wrong directory during cloning operation . . . . . . . Thick provisioning a large LUN with a wrapper volume fails with insufficient space error . . © Copyright IBM Corp. 2013 . . . . . . . . . . . . . . . . . . . . . . . 13 13 14 14 15 16 16 16 17 17 17 18 18 18 18 19 19 20 20 21 21 21 22 22 . . . . . . . 23 23 24 24 25 26 26 iii Datastore creation in Provisioning and Cloning stops and is rolled back due to sis start failure . . Error "TreeModel fails to return due to server-side errors" seen during cloning operations . . . . Using Provisioning and Cloning mount operation across data centers fails . . . . . . . . . Provisioning and Cloning fails to discover storage systems if licensed_feature.iscsi.enable=off . . Wrong disk format displayed after cloning a thin disk virtual machine on an NFS datastore . . . Fibre Channel datastore mount failure during Backup and Recovery mount and unmount processes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cloning fails with move error in log . . . . . . . . . . . . . . . . . . . . . . "Clone operation in progress" error after canceling task . . . . . . . . . . . . . . . Hardware compatibility error . . . . . . . . . . . . . . . . . . . . . . . . VM creation progress indicators do not match . . . . . . . . . . . . . . . . . . Storage controller data can take several seconds to load . . . . . . . . . . . . . . . Canceled tasks are not deleted from Recent Tasks pane . . . . . . . . . . . . . . . Misaligned VM file system impacts performance . . . . . . . . . . . . . . . . . Datastore Destroy operation fails . . . . . . . . . . . . . . . . . . . . . . . Performance is degraded when requesting large number of clones. . . . . . . . . . . . Password in custom specification not accepted . . . . . . . . . . . . . . . . . . VMs are not created . . . . . . . . . . . . . . . . . . . . . . . . . . . VMs remain after Destroy operation . . . . . . . . . . . . . . . . . . . . . . Multiple IPs are used on storage system . . . . . . . . . . . . . . . . . . . . Connectivity to storage controller is lost . . . . . . . . . . . . . . . . . . . . Redeploy fails on new virtual hard drive . . . . . . . . . . . . . . . . . . . . NFS error during Redeploy . . . . . . . . . . . . . . . . . . . . . . . . . Do not use e0M port . . . . . . . . . . . . . . . . . . . . . . . . . . . Internet script error during Redeploy . . . . . . . . . . . . . . . . . . . . . vFiler unit restrictions are ignored if physical storage system is added with root privileges . . . Known issues with Optimization and Migration . . . . . . . . . . . . . . . . . . . Scan Manager displays error message stating that datastores failed to load . . . . . . . . . Unlisted VMDK snapshot results in virtual machine going into Other folder . . . . . . . . Optimization and Migration screen sometimes needs a manual update . . . . . . . . . . Optimized datastores incorrectly appear in misaligned folder if no storage credentials are supplied Display of virtual machines or datastores sometimes stops briefly . . . . . . . . . . . . Virtual machines spanning multiple datastores incorrectly listed as misaligned . . . . . . . Known issues with Backup and Recovery . . . . . . . . . . . . . . . . . . . . . Error occurs during SnapManager for Microsoft SQL backup operation on Microsoft SQL databases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Backup and Recovery fails to enumerate the NFS exports when using an alias IP address . . . . Insufficient privileges error occurs and Backup and Recovery incorrectly reports the error . . . . Attempting to create an on-demand backup job to back up only the virtual machine fails with an error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Discovery of FC and iSCSI datastores might fail if the FC or iSCSI feature is not enabled on some hardware platforms . . . . . . . . . . . . . . . . . . . . . . . . . . . Restoring a VMFS virtual machine causes loss of independent disks . . . . . . . . . . . VMDKs should be removed from original datastore . . . . . . . . . . . . . . . . List of available virtual entities does not change if two datacenters have the same name . . . . Restoring a VMDK to an alternate datastore causes loss of the original VMDK on a virtual machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Virtual machine cannot power on after NFS or VMFS datastore is restored . . . . . . . . . Unable to delete restore session even if source virtual machine has been deleted . . . . . . . Successful backup job might trigger a notification during backup . . . . . . . . . . . . iv IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes 26 27 27 27 28 28 28 29 29 29 30 30 30 30 31 31 31 31 31 32 32 32 32 33 33 33 33 34 34 34 34 35 35 37 37 38 38 38 39 39 39 39 40 40 40 Virtual machine is unresponsive during a backup . . . . . . . . . . . . . . . . . Running a script during backup results in an error . . . . . . . . . . . . . . . . . Backup and Recovery cannot verify whether the scripts ran successfully . . . . . . . . . vCenter Server displays an error message that virtual machine has two VMDKs with same UUID Automount is enabled by default . . . . . . . . . . . . . . . . . . . . . . . Unable to verify which disk is being restored. . . . . . . . . . . . . . . . . . . Virtual machine is unresponsive during a backup that uses a VMware snapshot . . . . . . . CLI times out during a backup . . . . . . . . . . . . . . . . . . . . . . . Unable to mount an NFS volume if the export policy for the parent volume and the mounted backup volume are not the same . . . . . . . . . . . . . . . . . . . . . . . Disconnecting the FC initiator used to mount a LUN clone from the ESX server results in the mounted backup being removed from the server . . . . . . . . . . . . . . . . . Mounted backup not automatically removed when the restore session expires . . . . . . . . Single file restore on a Windows Server 2008 sometimes fails to bring disk online . . . . . . Single file restore does not automatically support VMDKs with multiple partitions . . . . . . Attaching a .sfr file using Outlook Web Access (OWA) might fail . . . . . . . . . . . . Deleting a nonexistent restore session results in an error . . . . . . . . . . . . . . . Drive letters that are used for network mapped drives might be incorrectly displayed as available drive letters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Notification email for restore session contains an IPv6 address . . . . . . . . . . . . . Restore session fails when using vFiler root volume . . . . . . . . . . . . . . . . Unable to retrieve deleted backup in offline session of Restore Agent . . . . . . . . . . . Unable to download Restore Agent setup file using Internet Explorer on Windows 2003 . . . . Known issues with the application programming interface for vCloud Director . . . . . . . . 40 40 40 41 41 41 41 42 42 42 42 43 43 43 44 44 44 44 45 45 45 Fixed issues . . . . . . . . . . . . Fixed issues: Monitoring and Host Configuration Fixed issues: Provisioning and Cloning . . . . Fixed issues: Optimization and Migration . . . Fixed issues: Backup and Recovery . . . . . . . . . . . . . . . . . . . . . . . 47 . . . . . . . . . . . . . . . . . . 47 . . . . . . . . . . . . . . . . . . 47 . . . . . . . . . . . . . . . . . . 48 . . . . . . . . . . . . . . . . . . 49 Where to find additional information . . . . . . . . . . . . . . . . . . . . . . 51 Websites . . . . . . . . . . . . . . . . . . . . . 53 . . . . . . . . . . Copyright and trademark information . . . . . . . . . . . . . . . . . . . . . 55 Trademark information . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contents . 59 v vi IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Virtual Storage Console 4.2 for VMware vSphere release notes Virtual Storage Console 4.2 for VMware vSphere consists of four capabilities: Monitoring and Host Configuration, Provisioning and Cloning, Optimization and Migration, and Backup and Recovery. These capabilities enable you to work with VMware and N series storage systems. The installation instructions and information about using each of the capabilities in Virtual Storage Console for VMware vSphere are in the Virtual Storage Console 4.1 for VMware vSphere Installation and Administration Guide. Version numbers for Virtual Storage Console for VMware vSphere Although this is the 4.2 release of Virtual Storage Console for VMware vSphere, the different components of Virtual Storage Console for VMware vSphere might use different numbering schemes internally. The following table summarizes the internal version numbers for individual components of Virtual Storage Console 4.2 for VMware vSphere: Virtual Storage Monitoring Console and Host version Configuration Optimization and Migration Provisioning and Cloning Backup and Recovery 4.2 1.2 4.2 5.1 4.2 System requirements for Virtual Storage Console for VMware vSphere Before you install the Virtual Storage Console for VMware vSphere software, you must ensure that your storage system conforms to the software, hardware, and operating system requirements. For the most up-to-date information on the system requirements, see the IBM N series interoperability matrix, available on the IBM N series support website, which is accessed and navigated as described in Websites on page 53. © Copyright IBM Corp. 2013 1 Updates to the Virtual Storage Console 4.2 for VMware vSphere release notes The release notes are updated between product releases as new information is uncovered. It is a good practice to check the release notes on a regular basis so that you can find out about any changes to the software. May 2013—The release notes were updated with the release of Virtual Storage Console 4.2 for VMware vSphere. The following information was added: v The standard Data ONTAP vsadmin role does not have sufficient privileges for Virtual Storage Console for VMware vSphere tasks. See “Data ONTAP vsadmin role insufficient for VSC for VMware vSphere tasks” on page 14. v If you install the NFS VAAI Plugin on an ESX host using NFS storage mounted from a direct virtual server (Vserver) connection, the status of the Vserver changes to unmanaged in Monitoring and Host Configuration, even though the plug-in installation is successful. See “Status of directly connected Vserver incorrectly reported in Monitoring and Host Configuration” on page 19. v When you perform a Virtual Storage Console for VMware vSphere cloning operation using Citrix XenDesktop as the connection broker, the operation places the XenDesktop export file in the wrong directory. See “XenDesktop export file placed in wrong directory during cloning operation” on page 26. v When provisioning a VMFS datastore with a wrapper volume, the datastore creation fails because the containing volume is not large enough. See Thick provisioning a large LUN with a wrapper volume fails with insufficient space error. v Efforts to use Provisioning and Cloning to create a deduped datastore fail when there is a busy sis schedule. See Datastore creation in Provisioning and Cloning stops and is rolled back due to sis start failure. v The error, TreeModel fails to return due to server-side errors, has been seen when users perform a task such as using the Create Rapid Clones wizard, which is provided by VSC. See Error "TreeModel fails to return due to server-side errors" seen during cloning operations. v The Provisioning and Cloning mount operation sometimes fails to perform correctly in environments that have multiple ESX hosts on different data centers on the same vCenter Server. See Using Provisioning and Cloning mount operation across data centers fails. v Optimization and Migration sometimes displays an error message on the Scan Manager panel, stating that it failed to load the datastores. See Scan Manager displays error message stating that datastores failed to load. v An error occurs during a SnapManager for Microsoft SQL backup operation on Microsoft SQL databases stored on virtual disks (VMDKs) or during an 2 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes v v v v unmount operation of a backup copy using Backup and Recovery. See Error occurs during SMSQL backup operation on Microsoft SQL databases. If you use an alias IP address, such as when an NFS volume is exported to the vCenter Server, Backup and Recovery does not list the storage entities when you access the storage system. See Backup and Recovery fails to enumerate the NFS exports when using an alias IP address. Attempting to create a backup job fails on a storage system for which a user does not have sufficient Data ONTAP privileges. This error is not reported correctly in the Backup and Recovery GUI. See Insufficient privileges error occurs and Backup and Recovery incorrectly reports the error. In Backup and Recovery, when you specify a one-time backup job to back up only the virtual machine and not the entire datastore, the backup job fails with an error. See Attempting to create an on-demand backup job to back up only the virtual machine fails resulting in an error. If you have a vNIC connected to an external Organization Virtual Data Center (Org vDC) network, cloning and provisioning operations default to vCloud Director operations. See “vApp operation defaults to vCloud Director operation if vNIC connected to organization vDC network” on page 45. v When performing a cloning operation on a vAPP that contains an NFS datastore, VSC writes invalid error messages to the kamino.log file. See “Invalid error messages seen when cloning a vAPP with an NFS datastore ” on page 46. Virtual Storage Console 4.2 for VMware vSphere release notes 3 4 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes New and changed features Virtual Storage Console 4.2 for VMware vSphere is an upgrade from Virtual Storage Console 4.1 for VMware vSphere and adds several new features. New features include the following: v Virtual Storage Console for VMware vSphere has added support for both vCenter Server and Data ONTAP role-based access control (RBAC). RBAC enables administrators to set up access for users so that they either have or do not have access to specific VSC tasks or vSphere objects. Note: The VSC-specific View privilege, which is read-only, is required for a user to view the VSC GUI. Without this privilege, menus, tabs, and other elements of the GUI are not visible. If you do not have this privilege and click the IBM icon on Home View, you get an error message. v You can use a set of VSC-specific, standard vCenter Server roles for working with VSC tasks. These roles provide the necessary vCenter native privileges and VSC-specific privileges to enable users to perform standard VSC tasks. v Monitoring and Host Configuration now manages storage discovery and credentials for the entire VSC product. All VSC capabilities now leverage the credentials entered during the initial setup and discovery process in Monitoring and Host Configuration, which eliminates the need for you to manage separate sets of storage entities in Backup and Recovery and unifies the experience within VSC. v VSC provides support for online alignment and migration of NFS datastores. This expanded support provides enhanced I/O performance, with online optimization and migration of misaligned virtual machines in NFS datastores (for Data ONTAP 8.1.3 and later). v By default, in the Backup wizard, the option to create a VMware snapshot is not selected for each backup copy. You must enable this option to trigger a VMware snapshot for each virtual machine during a backup. VSC for VMware vSphere and other products Virtual Storage Console for VMware vSphere supports additional products and features. In some cases, you must install these products separately or adjust your VSC environment to support a feature. © Copyright IBM Corp. 2013 5 Information on QFull settings in vSphere 5.1 vSphere 5.1 handles Task Set Full (QFull) conditions differently from vSphere 4.x and 5.0. You must set the QFull settings for each device. Information on the NFS VAAI Plugin The NFS VAAI Plugin does not ship with VSC for VMware vSphere. You must download the plugin from the NetApp Support Site and manually install it. 6 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Virtual Storage Console for VMware vSphere limitations To use VSC for VMware vSphere most efficiently, you should be aware that its performance can be affected by known issues with a particular feature, such as a network, or by features that a VSC for VMware vSphere capability does not support, such as a specific operating system version. You should also know that some of these limitations have workarounds that you can use. v “High-latency networks affect time required for rendering” v “VSC for VMware vSphere halts when installed on vCenter Server managing multiple ESX/ESXi hosts” v “Limitations that affect Monitoring and Host Configuration” on page 8 v “Limitations that affect Provisioning and Cloning” on page 9 v “Limitations that affect the Optimization and Migration capability” on page 10 v “Limitations that affect the Backup and Recovery capability and Restore Agent” on page 11 High-latency networks affect time required for rendering High-latency networks take longer to render objects. The rendering time is also affected by the number of vSphere clients and whether these clients are on the same machine as the vCenter Server. For example, suppose you want to load SAN details on 300 remote vSphere clients over a high-latency network. This task can take several minutes. Workaround: There is no workaround because this is normal behavior for high-latency networks. VSC for VMware vSphere halts when installed on vCenter Server managing multiple ESX/ESXi hosts Installing VSC for VMware vSphere on a vCenter Server that is managing a large number of ESX/ ESXi hosts can cause VSC for VMware vSphere to stop working properly. What happens is that the vCenter Server consumes all the CPU in order to manage the ESX/ESXi hosts. As a result, the vSphere Plugin Framework service cannot respond to requests from the VSC for VMware vSphere capabilities. This causes the capabilities to stop working. © Copyright IBM Corp. 2013 7 Workaround: Make sure that VSC for VMware vSphere has sufficient CPU and memory resources to run with the number of ESX/ESXi hosts in your environment. The amount of memory and resources required varies based on your system setup. Limitations that affect Monitoring and Host Configuration There are several limitations that affect Monitoring and Host Configuration. mbralign is not supported on Windows Server 2008 and above The mbralign program is not supported on Windows Server 2008, Windows Server 2008 R2, or Windows Server 2008 R2 SP1. By default, these versions of Windows Server 2008 are configured with GPT. Because of the offset of the file system, these systems normally do not need to be aligned. Note: In this situation with mbralign, the term Windows Server 2008 is used to refer to Windows Server 2008, Windows Server 2008 R2, and Windows Server R2 SP1. Workaround: If you use mbralign with Windows Server 2008, errors might occur when you boot the disk. So you must restore the Windows Server 2008 virtual machine by using the backup files that mbralign created. The mbralign program creates backup *-flat.vmdk and *.vmdk files in the same directory as the source .vmdk file. mbralign fails to align on Windows Server 2003 64-bit systems The mbralign program cannot align files on Windows Server 2003 64-bit versions. The program does not work with the NTFS 64-bit boot sectors. Workaround: You can restore the Windows Server 2003 64-bit VM by using the backup file that the mbralign program created. Data collection fails with "unknown error" on Windows Server 2008 and 2008 R2 systems When running the data collection programs with VSC for VMware vSphere installed on a Windows Server 2008 or Windows Server 2008 R2 system, you get an "unknown error" message. In this situation, the system displays the following error message: unknown error has occurred please see logs for details 8 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes The VSC for VMware vSphere service typically runs under the Local System account. For Windows Server 2008 and Windows Server 2008 R2, the Windows security features prevent the data collection programs from running correctly under the Local System account. Workaround: If you receive the preceding error message when running the data collection programs, configure the VSC for VMware vSphere service to run using an Administrator account. In the Windows Server 2008 or Windows Server 2008 R2 Services application, change the Log On properties of the VSC for VMware vSphere service to run as an Administrator account instead of the Local System account. Host discovery fails when VMware DPM powers down servers Monitoring and Host Configuration discovers only hosts that are connected and powered on. If hosts are powered down by VMware Distributed Power Management, Monitoring and Host Configuration does not discover them. Be sure all hosts are powered on and not in standby before host discovery begins. Limitations that affect Provisioning and Cloning There are several limitations that affect Provisioning and Cloning. Virtual machines with FaultTolerance enabled cannot be cloned The Provisioning and Cloning capability cannot clone virtual machines that have FaultTolerance enabled. This type of cloning operation is not supported by vCenter Servers. The Provisioning and Cloning capability follows the vCenter Server guidelines for cloning. NTFS partitions on dynamic disks not detected during space reclamation operation The Reclaim Space feature provided by Provisioning and Cloning does not support dynamic disks. If you execute this feature on a virtual machine that has NTFS partitions on a dynamic disk, it incorrectly reports that there are no NTFS partitions. Virtual Storage Console for VMware vSphere limitations 9 Cannot mount NFS datastores created using the -actual parameter The Provisioning and Cloning capability does not support mounting of NFS datastores that have been exported using the -actual parameter. VMware vSphere 5.x datastore limits do not apply in mixed vSphere environments In mixed-version vSphere environments, the maximum VMFS datastore size and maximum number of NFS datastores revert to vSphere 4 limits. Qtree Redeploy limitations The Provisioning and Cloning capability does not support redeploying qtrees in certain cases. The Redeploy feature is not supported in the following cases: v On qtrees residing on vFiler volumes v From NTFS qtrees to mixed qtrees Limitations that affect the Optimization and Migration capability At this time, there are some limitations that affect the Optimization and Migration capability of VSC for VMware vSphere. Optimization and Migration issues seen when FaultTolerance enabled Several Optimization and Migration issues have been found when working with virtual machines that have FaultTolerance enabled. If a virtual machine has FaultTolerance enabled, Optimization and Migration cannot perform the following tasks: v Scan the virtual machine when it is powered on and residing on a VMFS datastore. v Migrate the virtual machine. If you are using vSphere 5.1 and you attempt to migrate a virtual machine with FaultTolerance enabled, the migration fails, and the associated task remains incomplete. You must restart the vCenter service to clear this task. In addition, Optimization and Migration lists the primary and secondary virtual machines in a FaultTolerance configuration. 10 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Time required for scanning datastores increases in large environments If you have a large number of virtual machines, the Optimization and Migration capability scan operation can take a long time. For example, in tests using 1,371 virtual machines, three ESX hosts, and 140 datastores, the scan took 40 minutes. When scanning virtual machines, the Optimization and Migration capability uses VMware snapshots. Taking these snapshots increases the time required for the scan. As a result, as more virtual machines are scanned, the longer the scan takes. It is a good practice to schedule scans for non-critical production times. Limitations that affect the Backup and Recovery capability and Restore Agent Before you back up and restore datastores or virtual machines, you must be aware that you cannot use Backup and Recovery in some cases; this applies to Restore Agent as well. Following are some restrictions to using Backup and Recovery: v Backup and Recovery does not support initiating a multipath SnapMirror configuration from a backup. v v v v However, because Backup and Recovery does support a single-path SnapMirror initiation from a backup, you can use a multipath SnapMirror configuration along with Backup and Recovery if you ensure that the SnapMirror process occurs on a frequent schedule and is triggered from the storage system rather than from within Backup and Recovery through the SnapMirror job option. Backup and Recovery does not back up traditional volumes; it only backs up FlexVol volumes. Backup and Recovery does not restore a datastore that has been removed from vCenter Server after a backup. Backup and Recovery cannot back up a virtual machine that contains a disk in multi-writer mode.In such instances, VMware snapshots are not supported and selecting the Perform VMware consistency snapshot option for a backup fails. Backup and Recovery cannot mount VMFS-5 file system backups on ESX 4.x hosts. v Backup and Recovery does not support multiple vCenter Servers simultaneously. Virtual Storage Console for VMware vSphere limitations 11 v Backup and Recovery does not support datastores spanning more than one volume. v Backup and Recovery does not support VMFS datastores created with two or more LUN extents. v Backup and Recovery does not support more than four SCSI controllers for each virtual machine,and each controller can hold only up to 15 disks. If the virtual machine already has 15 disks on all four controllers, you cannot add more disks. v Backup and Recovery cannot take VMware snapshots of Fault Tolerance virtual machines. v Backup and Recovery cannot take VMware snapshots of Windows virtual machines that have iSCSI LUNs connected through Microsoft iSCSI Software Initiator or that have NPIV RDM LUNs. You can back up such virtual machines without selecting the Perform VMware consistency snapshot option. However, raw device maps are not backed up and are not recoverable from these backups. For more information, see VMware KB article 1009073: Unable to take a quiesced VMware snapshot of a virtual machine. In addition to the limitations that affect Backup and Recovery, there are also limitations that affect the Restore Agent, which works with Backup and Recovery during the restore process. The Restore Agent limitations include the following: v It does not support the mounting of GPT partitioned disks on Windows XP. v It does not support dynamic disks. v It does not support IDE controllers. 12 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Known issues You must be aware of certain known issues and workarounds for Virtual Storage Console 4.2 for VMware vSphere so that you can use the product more effectively. Some of these issues apply to all capabilities, while others apply to only a single capability. v “Known issues with VSC for VMware vSphere capabilities” v “Known issues with Monitoring and Host Configuration” on page 19 v “Known issues with Provisioning and Cloning” on page 25 v “Known issues with Optimization and Migration” on page 33 v “Known issues with Backup and Recovery” on page 35 v “Known issues with the application programming interface for vCloud Director” on page 45 Known issues with VSC for VMware vSphere capabilities VSC for VMware vSphere capabilities has the following known issues: v “Data ONTAP vsadmin role insufficient for VSC for VMware vSphere tasks” on page 14 v “No IP address found if VMkernel and storage systems not in same network” on page 14 v “VSC for VMware vSphere log export and file downloads fail when using IE 9” on page 15 v “Volume container not destroyed even if the last LUN had been deleted” on page 16 v “VSC for VMware vSphere fails to establish a secure connection with storage system” on page 16 v “Restarting the service delays start of VSC for VMware vSphere for 90 seconds” on page 16 v “Application configuration error” on page 17 v “Out of memory or script errors” on page 17 v “Error message "permission denied" when accessing the vSphere Client” on page 17 v “Clicking several links concurrently in the navigation pane results in an error” on page 18 v “Error message "A script on this page is causing Internet Explorer to run slowly..."” on page 18 © Copyright IBM Corp. 2013 13 v “Problem occurs when starting the vSphere Client if Virtual Storage Console server is unavailable” on page 18 v “Security certificate warnings displayed” on page 18 Data ONTAP vsadmin role insufficient for VSC for VMware vSphere tasks The standard Data ONTAP vsadmin role does not have sufficient privileges for Virtual Storage Console for VMware vSphere tasks. Instead of using the vsadmin role, you should create a new role using the RBAC User Creator for Data ONTAP tool, which is posted on the NetApp Communities Forum at: https://communities.netapp.com/docs/DOC-19074 No IP address found if VMkernel and storage systems not in same network Provisioning and Cloning and Optimization and Migration tasks might end in error if at least one VMkernel IP address and one storage system IP address are not in the same network. If kernel and storage system IP addresses are not in the same network, the following error is logged: Unable to continue. No ip addresses found when cross-referencing kernel ip addresses and addresses on the controller. Unable to find a matching network to NFS mount volume to these hosts. Workaround: Perform either of the following tasks: v Make sure that at least one VMkernel IP address and one storage system IP address are in the same network. v To allow mounting across all subnets or to specify allowed networks, use the following settings in the preferences file. The values in the preferences file are the ESX host subnet masks, not the controller subnet masks. These settings allow only the specified networks to map NFS datastores or iSCSI-based VMFS datastores, respectively: default.allow.nfs.mount.networks default.allow.iscsi.mount.networks In the following example, all NFS networks that share a common subnet can be used to mount datastores: 14 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes In the following example, only the 192.168.51.0 and 10.11.0.0 networks can map iSCSI-based VMFS datastores: There is a separate preferences file for Provisioning and Cloning and for Optimization and Migration. These files have the following default locations: Note: [VSC_home] is your VSC for VMware vSphere installation directory. v Provisioning and Cloning [VSC_home]\etc\kamino\kaminoprefs.xml v Optimization and Migration [VSC_home]\etc\caster\kaminosdkprefs.xml VSC for VMware vSphere log export and file downloads fail when using IE 9 In some cases, exporting VSC for VMware vSphere log files fails. This problem occurs when you are running Internet Explorer 9 with the Do not save encrypted pages to disk option enabled. The following error messages are displayed: Unable to download export-logs from [your VSC host]. Unable to open this Internet site. The requested site is either unavailable or cannot be found. Please try again later. This issue affects all versions of VSC for VMware vSphere. Workaround: There are two ways to resolve this issue. IBM recommends that you first try Method 1. If that does not work, then try Method 2. Method 1 From the Tools menu in Internet Explorer 9, perform the following tasks: 1. Click Internet Options. 2. Click the Advanced tab. 3. Clear the check from the Do not save encrypted pages to disk check box in the Security area. 4. Click OK. This is the default Internet Explorer setting. Method 2 Known issues 15 If Method 1 does not work, then perform these following tasks: 1. Start the Registry Editor. 2. For a per-user setting, locate the following registry key: HKEY_CURRENT_USER\SOFTWARE\Microsoft\Windows\CurrentVersion\Internet Settings For a per-computer setting, locate the following registry key: HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\ Internet Settings 3. On the Edit menu, click Add Value, and then add the following value: "BypassSSLNoCacheCheck"=Dword:00000001 4. Exit the Registry Editor. Related information: Error message when you download files by using Internet Explorer 9 from secure websites: " couldn't be downloaded" Volume container not destroyed even if the last LUN had been deleted In some environments, VSC for VMware vSphere does not delete the volume container even though the last LUN has been deleted and the option "Delete a volume if the last LUN in it has been deleted" is enabled. This issue has been seen in environments using vFiler units with Data ONTAP operating in 7-Mode. Workaround: There is no workaround for this issue at this time. VSC for VMware vSphere fails to establish a secure connection with storage system VSC for VMware vSphere fails to connect to a storage system using SSL. This problem is seen when the storage system has the ssl.v2.enable option set to false because it is attempting to use SSLv3. This issue occurs because of the way Data ONTAP implements SSL. Workaround: To prevent this issue, you must make sure the ssl.v2.enable option is turned on. After VSC for VMware vSphere establishes a connection with the storage system, it then uses an SSLv3 connection. This issue is resolved in Data ONTAP 8.0.4 and 8.1.1. Restarting the service delays start of VSC for VMware vSphere for 90 seconds When you restart the Virtual Storage Console for VMware vSphere service, you must wait 90 seconds before you can start vSphere Client. 16 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes If you start the vSphere Client before the 90 seconds has completed, you get a message saying the connection failed. The VSC for VMware vSphere plug-in is displayed as disabled. After the 90 seconds has passed, you can launch the vSphere Client without any problem. This issue is connected to the disk I/O speed and the amount of time it takes vSphere Plugin Framework to unpack the VSC for VMware vSphere .war files. Application configuration error You may see an error similar to the following when using the Provisioning and Cloning capability to clone a virtual machine from non-N series or VMFS datastores, or when trying to import virtual machines into VMware View. It may also occur in the Optimization and Migration capability when scanning non-N series or VMFS datastores. This application has failed to start because the application configuration is incorrect. Reinstalling the application may fix this problem. Workaround: Run Windows updates to install the latest C++ runtime libraries. Out of memory or script errors When VSC for VMware vSphere displays inventory items in a new modal window, vCenter allocates a new instance of an Internet Explorer window. When you apply an operation or cancel the current one, VSC for VMware vSphere dismisses the window but cannot clear the memory consumed by the vpxclient process. For example, if you open and cancel the destroy datastore feature several times, the vpxclient process memory consumption grows, but does not shrink, even when the window is dismissed. Workaround: Restart the vSphere Client. Error message "permission denied" when accessing the vSphere Client After installing VSC for VMware vSphere, you might see this message when you launch the vSphere Client to access VSC for VMware vSphere because of the proxy configuration in Internet Explorer. Workaround: Disabling the Internet Explorer proxy configuration enables access to VSC for VMware vSphere using the vSphere Client. To turn off the proxy settings, complete the following steps: Known issues 17 1. 2. 3. 4. Close any vSphere Clients that are open. Navigate to the Control Panel. Click Internet Options and select the Connections tab. Click LAN Settings. 5. Clear all the check boxes and click OK. After you disable the proxy settings, launch the vSphere Client and click the IBM N series tab to verify that the error message does not appear. Clicking several links concurrently in the navigation pane results in an error If you click several links concurrently in the vSphere Client navigation pane, access to items in the pane might fail with the following error message: An error has occurred in the script on this page Workaround: If this error message appears, restart the vSphere Client. To avoid this problem, click another link in the navigation pane only after the current panel appears. Error message "A script on this page is causing Internet Explorer to run slowly..." You might see this message if VSC for VMware vSphere takes too long to complete an operation. You can adjust your Internet Explorer settings to avoid this message. This message does not indicate a problem with VSC for VMware vSphere other than a script taking longer to run than the timeout setting in Internet Explorer. For instructions on adjusting the Internet Explorer timeout setting, see Microsoft KB article 175500: Error message: "A script on this page is causing Internet Explorer to run slowly". Problem occurs when starting the vSphere Client if Virtual Storage Console server is unavailable If the Virtual Storage Console server is unavailable when starting the vSphere Client, the IBM N series icon does not appear. Workaround: Restart the vSphere Client when the Virtual Storage Console server is available. Security certificate warnings displayed Depending on your web browser and security configuration, you might see security certificate warnings displayed when installing and using VSC for 18 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes VMware vSphere. You can safely ignore these warnings and proceed to the requested program or web page. VSC for VMware vSphere uses self-signed SSL certificates. You might see warnings similar to the following: The security certificate presented by this website was not issued by a trusted certificate authority. You can also add self-signed certificates or a certificate signed by a Certificate Authority. Known issues with Monitoring and Host Configuration There are several known issues that apply to the Monitoring and Host Configuration capability. v “Status of directly connected Vserver incorrectly reported in Monitoring and Host Configuration” v “ESXi might finish alignment with different LBAs than stated with the -preview option” on page 20 v “mbralign for ESXi takes a long time to align files” on page 20 v “VM does not boot after running mbralign” on page 21 v “MPIO Settings still shows Alert icon after running Set Recommend Values” on page 21 v “Windows cluster disk reservation lost after faults in ALUA configuration” on page 21 v “Windows guest OS might crash during storage faults because of virtual memory use” on page 22 v “ESX with Windows Server 2003 MSCS guest uses partner path after controller giveback” on page 22 v “ESX or ESXi 4 in an FC or FCoE configuration with ALUA might show all dead paths for some LUNs after a controller fault” on page 23 v “Virtual machine disk I/O throughput can drop without sufficient physical memory” on page 23 v “Cisco 10-Gb network or FCoE switch port in error disabled state” on page 24 v “vSphere Client shows inactive datastores for ESX/ESXi hosts” on page 24 Status of directly connected Vserver incorrectly reported in Monitoring and Host Configuration When using Virtual Storage Console for VMware vSphere to install the NFS VAAI Plugin on an ESX host using NFS storage mounted from a direct virtual Known issues 19 server (Vserver) connection, the status of the Vserver changes to unmanaged in Monitoring and Host Configuration, even though the plug-in installation is successful. Workaround: You must connect to the storage controller using a cluster-management LIF by manually typing the nfs modify -vserver vServer_name -vstorage enabled command for the appropriate Vserver, and then click the Update link in Monitoring and Host Configuration to refresh the Vserver status to its normal state. ESXi might finish alignment with different LBAs than stated with the -preview option When you use mbralign for ESXi, the resulting LBAs might be different from the ones stated with the -- preview option. This difference does not prevent the alignment from occurring or pose a problem for the virtual machines. The ESXi version of mbralign is based on the mbralign from ESX Host Utilities 5.1. The ESX version was updated to include bug fixes after the ESXi version was created. The ESXi version of mbralign successfully performs alignments, so it has not been updated even though the --preview option display might not be correct. Workaround: There are two ways to handle this issue: v Use mbralign for ESX. Migrate the virtual machines from the ESXi server to the ESX server. Align them. Then migrate them back. v Continue to use mbralign for ESXi and note that the resulting LBAs might be slightly different. mbralign for ESXi takes a long time to align files The mbralign program for ESXi can take a much longer time to align files than the mbralign program used with the ESX Host Utilities 5.2.1. This happens because the mbralign program runs in the technical support console of ESXi, which is a busybox console. As a result, the number of resources available for the application are reduced. The limited amount of memory causes an increase in the amount of time it takes to align the VMDK files. Workaround: There is no workaround at this time because there currently is no way to increase the amount of memory in the technical support console of ESXi. 20 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes VM does not boot after running mbralign In certain situations, a virtual machine might not boot after running the mbralign program. You can revert to the backup VMDK files saved by mbralign. There are several problems that can prevent a VM from booting after running the mbralign program. The workaround, however, is the same: revert to the backup VMDK files saved by the mbralign program. Avoid the following situations when running the mbralign program: v Do not run mbralign for a VM when the VM is powered on. v Do not run mbralign on SLES 11 guests. v Do not use the --force option when running the mbralign program on Windows 7, Windows Server 2008, or Windows Server 2008 R2 guests. MPIO Settings still shows Alert icon after running Set Recommend Values After running Set Recommended Values, the MPIO Settings column for an ESX host on the Monitoring and Host Configuration capability Overview panel might still show an Alert icon. If a storage controller is skipped during discovery, the Monitoring and Host Configuration capability is unable to set the MPIO settings for the paths to that controller. The Alert icon continues to be displayed after running Set Recommended Values. Workaround: You can check the Discovery Status panel to see if you have any skipped controllers. If you have a skipped controller, right-click the controller and select Modify Credentials. Clear the Skipped check box and enter credentials for the controller. Run an Update, and then run Set Recommended Values again. Windows cluster disk reservation lost after faults in ALUA configuration A Windows cluster (Windows Server 2003 MSCS or Windows Server 2008 failover cluster) running as a guest on ESX 4.0 reports disk reservations lost. The cluster service is unavailable and delayed write errors might be reported. The LUNs used by the cluster have ALUA enabled. This problem can occur with all ESX path selection policies. Workaround: When using ESX guest operating systems in a Windows cluster configuration, disable ALUA on the LUNs and use the FIXED path selection policy. Known issues 21 Windows guest OS might crash during storage faults because of virtual memory use Windows guest operating systems on ESX 4.0 servers might crash during a storage fault due to a kernel in page stack error because the virtual memory on a SAN disk is temporarily unavailable. The Windows event log might contain errors similar to the following: Event ID: 51 Event Type: Warning Event Source: Disk Description: An error was detected on device \Device\Harddisk0\DR0 during a paging operation. Event ID: 9 Source: %HBA_DRIVER_NAME% Description: The device, \Device\ScsiPort0, did not respond within the timeout period. Workaround: There are several possible workarounds you can try, including the following: v Allocating sufficient physical memory to the guests v Placing the pagefile on local disk v Minimizing the virtual memory size on the Windows guests where possible to minimize its usage Note that having the pagefile on local disk does not allow vMotion on the guest to work properly, because vMotion requires a shared disk. ESX with Windows Server 2003 MSCS guest uses partner path after controller giveback After controller giveback, the I/O path to a LUN used by a Windows Server 2003 MSCS guest continues to use a partner (proxy) path. The controller issues a partner path misconfigured message, and performance might be degraded. This issue occurs in a VMware ESX configuration with two Windows Server 2003 MSCS guests clustered across physical ESX hosts accessing shared disk resources from a LUN on an high-availability (HA) pair. Before the issue occurs, one controller successfully fails over to its partner. However, after controller giveback, the I/O path to the LUN continues to use a partner path. This issue occurs only when the controller that usually owns the LUN is the controller that failed over. During controller failover, all I/O to the failed controller switches to the partner controller. After giveback, the I/O should return to the original path through the original controller. If the I/O continues through the partner 22 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes controller, the routing is less efficient. The storage controller issues messages saying the partner path is misconfigured to notify you that I/O is using a partner path. In this MSCS configuration, ESX continues to use the partner path because that path has a SCSI reservation for the LUN. ESX uses the partner path to the LUN until the disk resource ownership is transitioned to the other VM or the current partner path becomes unavailable. The ESX host then selects a new path. By design, the ESX host returns to the initial preferred path that was specified for the LUN if it is available. Workaround: You can manually transition the disk resources to the ESX host where the other MSCS guest is located. The ESX host for that guest sends a SCSI RESERVE command to the disk, and ESX selects a local path to the LUN. ESX or ESXi 4 in an FC or FCoE configuration with ALUA might show all dead paths for some LUNs after a controller fault Running ESX 4 or ESXi 4 in an FC or FCoE configuration with ALUA, the ESX host might indicate that all paths to a LUN are dead after one of the storage controllers experiences a fault. This can happen in a High Availability configuration. This issue can occur if the ESX path evaluation initiated after the fault happens before the controller's ALUA state transition is completed. No action is needed. If there are valid physical paths to the LUN, the condition is cleared by either of the following events: v Periodic path evaluation occurs, which happens every 300 seconds. v An I/O is sent to the LUN. Either event enables the host to determine the proper ALUA state and update the path state. Virtual machine disk I/O throughput can drop without sufficient physical memory VMware virtual machine disk I/O throughput can drop significantly when memory usage on other virtual machines increases. This issue occurs when physical memory is overcommitted and ESX must use memory reclamation and swapping. The ESX and ESXi memory management techniques allow memory overcommitment on the host, which works well on typical deployments where the virtual machines are lightly loaded. When the working set memory in all virtual machines combined plus the virtualization overhead exceeds the physical memory available on the host, ESX or ESXi uses memory reclamation and swapping to provide additional memory needed for the applications. This Known issues 23 can impact the disk I/O performance. When predictable disk I/O performance is needed and when the host does not have sufficient physical memory, the memory reclamation impact in a virtual machine can be limited by reserving sufficient memory for that virtual machine. See the Managing Memory Resources chapter in the vSphere Resource Management Guide for additional information about memory virtualization and performance considerations. Cisco 10-Gb network or FCoE switch port in error disabled state After storage controller faults, the ports on a Cisco 10-Gb network or FCoE switch might go into error disabled (errdisable) state. The error disabled state is caused by too many flow-control (pause) frames being sent to the port. You can configure your switch to automatically recover from this problem. Cisco Ethernet switch ports respond to a large number of flow-control frames by disabling the port. This is intended to protect the switch from various failures and configuration problems. By default, after the port is disabled, it must be re-enabled manually. During storage controller faults, the number of flow-control frames sent to the switch can exceed the limit and cause the port to be disabled. Workaround: You can configure the switch to automatically recover the port from this state after a specified interval by using the following switch commands: errdisable recovery cause pause-rate-limit errdisable recovery interval 30 For more information about the errdisable command, see the documentation for the Cisco switch. vSphere Client shows inactive datastores for ESX/ESXi hosts After a storage fault, the vSphere Client might show one or more datastores for an ESX/ESXi 4.1, 4.1u1, or ESXi 5.0u1 host as inactive. VMware features that require the datastore to be active are not available. Features including VMware vMotion, VMware High Availability, VMware Distributed Resource Scheduler, and VMware Fault Tolerance are not available when the datastore status is inactive. The workaround is to rescan the storage adapters by selecting Storage Adapters on the Configuration tab in the vSphere Client and clicking Rescan. After the rescan, the vSphere Client shows the correct datastore status. 24 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Known issues with Provisioning and Cloning There are several known issues that apply to the Provisioning and Cloning capability. v “XenDesktop export file placed in wrong directory during cloning operation” on page 26 v “Thick provisioning a large LUN with a wrapper volume fails with insufficient space error” on page 26 v “Datastore creation in Provisioning and Cloning stops and is rolled back due to sis start failure” on page 26 v “Error "TreeModel fails to return due to server-side errors" seen during cloning operations” on page 27 v “Using Provisioning and Cloning mount operation across data centers fails” on page 27 v “Provisioning and Cloning fails to discover storage systems if licensed_feature.iscsi.enable=off” on page 27 v “Wrong disk format displayed after cloning a thin disk virtual machine on an NFS datastore” on page 28 v “Fibre Channel datastore mount failure during Backup and Recovery mount and unmount processes” on page 28 v “Cloning fails with move error in log” on page 28 v “"Clone operation in progress" error after canceling task” on page 29 v “Hardware compatibility error” on page 29 v v v v v “VM creation progress indicators do not match” on page 29 “Storage controller data can take several seconds to load” on page 30 “Canceled tasks are not deleted from Recent Tasks pane” on page 30 “Misaligned VM file system impacts performance” on page 30 “Datastore Destroy operation fails” on page 30 v “Performance is degraded when requesting large number of clones” on page 31 v “Password in custom specification not accepted” on page 31 v “VMs are not created” on page 31 v “VMs remain after Destroy operation” on page 31 v v v v v “Multiple IPs are used on storage system” on page 31 “Connectivity to storage controller is lost” on page 32 “Redeploy fails on new virtual hard drive” on page 32 “NFS error during Redeploy” on page 32 “Do not use e0M port” on page 32 v “Internet script error during Redeploy” on page 33 Known issues 25 v “vFiler unit restrictions are ignored if physical storage system is added with root privileges” on page 33 XenDesktop export file placed in wrong directory during cloning operation When you perform a Virtual Storage Console for VMware vSphere cloning operation using Citrix XenDesktop as the connection broker, the operation places the XenDesktop export file in the \etc\Kamino directory. The correct location for the file is \etc\Kamino\exports. Workaround: If you encounter this problem, you should manually move the XenDesktop export file to the \etc\Kamino\exports directory. The name for the XenDesktop export file always uses the format: xenDesktop_.csv Thick provisioning a large LUN with a wrapper volume fails with insufficient space error When provisioning a VMFS datastore with a wrapper volume, Virtual Storage Console for VMware vSphere sets the size of the wrapper volume to 5 GB larger than the datastore size. For LUNs 1 TB and larger, this is not a sufficient buffer, so the datastore creation fails because the containing volume is not large enough. Workaround: To resolve this issue, increase the buffer size for the volume by editing etc/ kaminoprefs.xml. The option that controls the buffer size is default.create.volume.lunSpaceBuffer. The value you supply is interpreted as GB. The default value is 5. Keep increasing this value until the provisioning operation succeeds. Note: Setting the default.create.volume.lunSpaceBuffer affects all wrapper volumes that VSC creates, regardless of the size of the LUN. Datastore creation in Provisioning and Cloning stops and is rolled back due to sis start failure Efforts to use Provisioning and Cloning to create a deduped datastore fail when there is a busy sis schedule. If this happens, the datastore creation operation is rolled back because of a sis start failure. Workaround: To avoid this problem, disable deduplication in Virtual Storage Console for VMware vSphere before you use Provisioning and Cloning to create a datastore. 26 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Error "TreeModel fails to return due to server-side errors" seen during cloning operations The error "TreeModel fails to return due to server-side errors" has been seen when performing a task such as using the Create Rapid Clones Wizard, which is provided by Virtual Storage Console for VMware vSphere. This error message indicates that the host is disconnected. It does not indicate a problem with VSC. Rebooting the vCenter Server when a host is disconnected causes the host to miss some required information in its internal state. This unexpected state then causes an unhandled error to occur when you try to perform an operation such as cloning. Workaround: If you encounter this issue, reconnect the host to the vCenter Server. Doing this resolves the internal state of the host object with vSphere. You can confirm the invalid state of the host using the vSphere Managed Object Browser. Navigate to the page for that host. If host is in an invalid state, the configManager property either will not have a value or the value will be Unset. Using Provisioning and Cloning mount operation across data centers fails The Virtual Storage Console for VMware vSphere Provisioning and Cloning mount operation sometimes fails to perform correctly in environments that have multiple ESX hosts on different data centers on the same vCenter Server. Instead of mounting the new datastores on a new ESX host, Provisioning and Cloning remounts the datastores currently mounted on one of the other hosts. If you continue to attempt the mount operation, it fails. Workaround: At this time, there is no workaround for this issue. Provisioning and Cloning fails to discover storage systems if licensed_feature.iscsi.enable=off If a storage system is running Data ONTAP operating in 7-Mode and the licensed_feature.iscsi.enable parameter is set to off instead of on, Provisioning and Cloning reports an exception and does not discover the storage controller. This problem only occurs with Provisioning and Cloning. Monitoring and Host Configuration is able to add the storage controller. Workaround: To enable Provisioning and Cloning to discover the storage controller, you must set licensed_feature.iscsi.enable to on. Known issues 27 Wrong disk format displayed after cloning a thin disk virtual machine on an NFS datastore The cloning feature of the Provisioning and Cloning capability incorrectly displays the disk format when you specify thick format for the clone, but the original virtual machine uses a thin disk format and is on an NFS datastore. The Provisioning and Cloning capability reports that the cloned virtual machine is using thin format. The operation actually reserves enough volume space for the thick disk format. It just reports the format of the cloned disk incorrectly. This issue is only seen when the original virtual machine uses thin disk format and resides on an NFS datastore. Fibre Channel datastore mount failure during Backup and Recovery mount and unmount processes If you attempt to mount an FC datastore using the Provisioning and Cloning capability while a mount or unmount is in progress by the Backup and Recovery capability, the datastore mount might fail. Workaround: Do not mount a datastore with the Provisioning and Cloning capability while performing a mount or unmount using the Backup and Recovery capability. Cloning fails with move error in log A cloning task might fail when the controller cannot resolve its host name or when a controller resolves its host name to an IP address other than one it owns. In either case, the ndmpcopy operation fails. When cloning fails, an error similar to the following error is logged: ERROR com.netapp.rcu.server.jzapi.FilerUtil - Move of /vol/win7_64bit_test/ ntap_rcu1268850273837/.tmpNdmpDir/Win7_64bitflat.vmdk to /vol/ win7_64bit_test/ntap_rcu1268850273837/0000hd0000-flat.vmdk failed. The following example shows a configuration in which the IP address 10.61.177.220 was removed from a controller and the etc/hosts file was not updated. The etc/hosts file contains both IP addresses. #Auto-generated by setup Thu Nov 19 15:47:12 GMT 2009 127.0.0.1 localhost 10.61.177.220 kc3140 <= (This is wrong) 10.61.177.221 kc3140-e0M The output of ifconfig -a shows that only the 10.61.177.221 address is operational on this controller: 28 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes kc3140> ifconfig -a e0M: flags=0x2948867 mtu 1500 inet 10.61.177.221 netmask 0xffffff00 broadcast 10.61.177.255 ether 00:a0:98:0a:22:06 (auto-100tx-fd-up) flowcontrol full Workaround: Ensure that the controller can resolve its name to a valid IP address. In this example, map 10.61.177.221 to kc3140: #Auto-generated by setup Thu Nov 19 15:47:12 GMT 2009 127.0.0.1 localhost 10.61.177.221 kc3140 "Clone operation in progress" error after canceling task After a clone operation is canceled and the canceled message is displayed in the Recent Tasks pane, a new clone operation request can result in an error. If the Web service has not finished cleaning up the canceled task before a new request is made, the following error is displayed: A clone operation is already in progress. Once it’s completed, try again. Workaround: Wait a few seconds after canceling a clone operation before starting a new one. Hardware compatibility error The destination ESX host, or the ESX hosts in the destination datacenter or cluster, are not checked for complete hardware compatibility prior to cloning. If the source and destination are not compatible, the cloning operation fails. When a cloning operation fails due to hardware incompatibility, an error is logged. For example, a request to clone a source virtual machine with four virtual CPUs to an ESX host with only one physical CPU would result in the following error: Virtual Machine requires 4 CPUs to operate, but the host hardware provides 1. Workaround: Ensure that the destination hardware is compatible with the source before executing a cloning operation. VM creation progress indicators do not match You can safely ignore the mismatch of progress indicators for virtual machine creation in vSphere Client panels. Known issues 29 Storage controller data can take several seconds to load The fields in the Specify Storage Controllers screen can take several seconds to load. The Provisioning and Cloning capability cross-references datastore and controller information while loading storage controller data. The delay varies depending on the sizes of the environment, the server running VSC for VMware vSphere, and the server running vCenter. Canceled tasks are not deleted from Recent Tasks pane Canceled Provisioning and Cloning tasks might remain displayed in the vSphere Client Recent Tasks pane indefinitely. Workaround: To clear these tasks, do one of the following: v Restart the Windows service 1. Select Control Panel > Administrative Tools > Services. 2. Right-click VMware vCenter Server and select Restart. v Run the following commands from the command line 1. net stop vpxd 2. net start vpxd Misaligned VM file system impacts performance If the file system in the virtual machine is not properly aligned, a warning is displayed before the cloning operation begins. This enables you to correct the misalignment and avoid propagating the misalignment to the clones. Messages are also logged in the messages file on the controller and in the AutoSupport data. Performance is significantly impacted by the block alignment of the virtual machine file system to the controller. The Provisioning and Cloning capability inspects the source VMDK file of any virtual machine that is cloned. The result of this inspection is recorded in the log file. If the file system in the virtual machine is not properly aligned, the following message is logged: Partition 1 of /vol/test/ntap_rcu1236650660984/0000hd0000-flat.vmdk is NOT aligned. (LBA=63). Partition number of volume_path-flat.vmdk is NOT aligned. Datastore Destroy operation fails During a datastore Destroy operation, if any host cannot disconnect the datastore, the operation fails. Failure of the Destroy operation can be caused by a virtual machine being powered up after the Destroy process is initiated. 30 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Workaround: Resolve the issue that caused the datastore disconnect to fail. Then perform the Destroy operation from the ESX host that cannot disconnect. Performance is degraded when requesting large number of clones Cloning performance is affected by many factors, including vCenter Server hardware configuration, number and hardware configuration of the ESX hosts, and the current load on either or both. Workaround: Performance can degrade if you request a very large number of clones in a single operation. If this occurs, reduce the number of clones requested per operation. For example, instead of requesting 2,000 clones, perform two operations of 1,000 clones each. Password in custom specification not accepted The password in a custom specification might not be applied. This occurs in Windows operating systems only. Workaround: Power up the source virtual machine, set the password in the source virtual machine, and power it down. Then rerun the cloning request. VMs are not created During storage system takeover or giveback, virtual machine clone creation fails because the Provisioning and Cloning capability cannot communicate with the storage controller. Workaround: Resubmit the cloning request when the takeover or giveback process is complete. VMs remain after Destroy operation After running the Destroy operation on a datastore, some of the VMs might still appear in the inventory tree. Attempting to remove these destroyed VMs results in the following error message: The request refers to an object that no longer exists or has never existed. Workaround: Restart the vSphere Client. Multiple IPs are used on storage system Any matching network between the storage system and the ESX VMkernel is used to mount datastores. To prevent a network from being used to mount datastores, use the following setting in the preferences file: Known issues 31 default.restrict.mount.networks The preferences file is stored by default in the following location, where [VSC_home] is your VSC for VMware vSphere installation directory: [VSC_home]\etc\kamino\kaminoprefs.xml In the following example, the networks 192.168.2.0 and 10.1.0.0 are ignored when determining the storage system IP address to use for the next datastore. Connectivity to storage controller is lost If connection to the storage controller is lost, Provisioning and Cloning tasks in progress will fail. Workaround: Resolve the communication issue. No additional steps need to be taken in the application. The Provisioning and Cloning capability creates a new connection for each new task. Redeploy fails on new virtual hard drive If you attempt a Redeploy operation after adding a new virtual hard drive to the source virtual machine rather than modifying the contents of the existing hard drives, the operation fails. When a new virtual hard drive is added to the virtual machine source after clones are created, the new hard drive does not exist on the clones. Therefore, there is no destination datastore identified for the new hard drive and the Redeploy operation fails. Workaround: After adding a new virtual hard drive to a source virtual machine, create new clones from the source rather than performing a Redeploy operation. NFS error during Redeploy The Redeploy operation fails if an ISO is mounted to the clones that are to be redeployed. Workaround: Remove the ISO mapping and run the operation again. Do not use e0M port When adding storage controllers for use with the Provisioning and Cloning capability, do not select the e0M port when specifying interfaces to use on the storage controller. 32 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Internet script error during Redeploy Occasionally, an internet script error displays after clicking the Redeploy link. Click OK on the error message and the procedure completes normally. vFiler unit restrictions are ignored if physical storage system is added with root privileges If a physical filer and its vFiler units are added to the Provisioning and Cloning capability, any restrictions on the vfiler unit are ignored if the physical storage system has root privileges. For example, a user added in vFiler context without destroy capabilities could destroy datastores if the vFiler unit's physical storage system was added with root privileges. Workaround: To restrict access to a vFiler unit, only the vFiler unit, and not the physical storage system, should be added to the Provisioning and Cloning capability. Known issues with Optimization and Migration There are several known issues that apply to the Optimization and Migration capability. v “Scan Manager displays error message stating that datastores failed to load”“Unlisted VMDK snapshot results in virtual machine going into Other folder” on page 34 v “Optimization and Migration screen sometimes needs a manual update” on page 34 v “Optimized datastores incorrectly appear in misaligned folder if no storage credentials are supplied” on page 34 v “Display of virtual machines or datastores sometimes stops briefly” on page 34 v “Virtual machines spanning multiple datastores incorrectly listed as misaligned” on page 35 Scan Manager displays error message stating that datastores failed to load The Scan Manager of Optimization and Migration sometimes displays an error message stating that it failed to load datastores. This happens when the time it takes to refresh the Scan Manager panel is longer than the default refresh interval of 15 seconds. This issue is normally seen in large environments. Workaround: To avoid this issue, increase the refresh rate to 60 seconds by completing the following steps: Known issues 33 1. In \etc\caster\casterprefs.xml, add the following entries: 2. Restart the VSC Windows service. Unlisted VMDK snapshot results in virtual machine going into Other folder If the Optimization and Migration capability encounters a virtual machine with an unlisted VMDK snapshot while performing a scan operation, it places that virtual machine in the Other folder. The Optimization and Migration utility does not support configurations with unlisted snapshots. Related information: VMware KB article 1005049: Determining if there are leftover delta files or snapshots that VMware vSphere or Infrastructure Client cannot detect Optimization and Migration screen sometimes needs a manual update In some cases, the Optimization and Migration screen fails to automatically update and must be manually updated. This issue has been seen only on the Virtual Machine Alignment panel when a node is selected in the folder tree. To manually update the page, click Refresh. Optimized datastores incorrectly appear in misaligned folder if no storage credentials are supplied The Optimization and Migration capability sometimes incorrectly puts optimized datastores in the misaligned folder. This issue occurs when you do not supply storage credentials for the IBM N series storage systems. This issue or similar issues where a datastore is placed in the wrong folder can also occur if you use non-IBM N series storage systems, which, by default, do not have IBM N series storage credentials. To avoid this problem, make sure you set up storage credentials for each storage system. See the VSC for VMware vSphere installation guide for information on setting up storage credentials. If you want to migrate datastores from non-IBM N series storage systems, you can use the Optimization and Migration capability to do this. After you have migrated the datastores, make sure they are on IBM N series storage systems that use storage credentials. Display of virtual machines or datastores sometimes stops briefly If you use the Optimization and Migration capability to scan a virtual machine running vCenter Server, you might not be able to see information about the virtual machines or datastores for a brief period of time. 34 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes This is a temporary problem that occurs when the following is true: v The virtual machine being scanned is running vCenter Server. v The virtual machine requires that a VMware snapshot be made. The snapshot occurs when the virtual machine is on VMFS or when the controller containing it cannot be reached. After being created, the snapshot is deleted, which interrupts communication with the network until VSC for VMware vSphere reestablishes communication. Virtual machines spanning multiple datastores incorrectly listed as misaligned In some cases, the Optimization and Migration capability places virtual machines in the Other folder even though they are aligned. The Optimization and Migration capability does this because it cannot access the virtual machines after scanning them. This situation is seen when virtual machines reside in multiple datastores and some of the datastores are on VMFS. As part of the scan process the Optimization and Migration capability creates a snapshot of the datastore. At this time there is no workaround for this issue. Known issues with Backup and Recovery The description of known issues in Backup and Recovery can help you to recognize issues and the corrective actions that are available to you to resolve the issues. v “Error occurs during SnapManager for Microsoft SQL backup operation on Microsoft SQL databases” on page 37 v “Backup and Recovery fails to enumerate the NFS exports when using an alias IP address” on page 37 v “Insufficient privileges error occurs and Backup and Recovery incorrectly reports the error” on page 38 v “Attempting to create an on-demand backup job to back up only the virtual machine fails with an error” on page 38 v “Discovery of FC and iSCSI datastores might fail if the FC or iSCSI feature is not enabled on some hardware platforms” on page 38 v “Restoring a VMFS virtual machine causes loss of independent disks” on page 39 v “VMDKs should be removed from original datastore” on page 39 v “List of available virtual entities does not change if two datacenters have the same name” on page 39 Known issues 35 v “Restoring a VMDK to an alternate datastore causes loss of the original VMDK on a virtual machine” on page 39 v “Virtual machine cannot power on after NFS or VMFS datastore is restored” on page 40 v “Unable to delete restore session even if source virtual machine has been deleted” on page 40 v “Successful backup job might trigger a notification during backup” on page 40 v “Virtual machine is unresponsive during a backup” on page 40 v “Running a script during backup results in an error” on page 40 v “Backup and Recovery cannot verify whether the scripts ran successfully” on page 40 v “vCenter Server displays an error message that virtual machine has two VMDKs with same UUID” on page 41 v “Automount is enabled by default” on page 41 v “Unable to verify which disk is being restored” on page 41 v “Virtual machine is unresponsive during a backup that uses a VMware snapshot” on page 41 v “CLI times out during a backup” on page 42 v “Unable to mount an NFS volume if the export policy for the parent volume and the mounted backup volume are not the same” on page 42 v “Disconnecting the FC initiator used to mount a LUN clone from the ESX server results in the mounted backup being removed from the server” on page 42 v “Mounted backup not automatically removed when the restore session expires” on page 42 v “Single file restore on a Windows Server 2008 sometimes fails to bring disk online” on page 43 v “Single file restore does not automatically support VMDKs with multiple partitions” on page 43 v “Attaching a .sfr file using Outlook Web Access (OWA) might fail” on page 43 v “Deleting a nonexistent restore session results in an error” on page 44 v “Drive letters that are used for network mapped drives might be incorrectly displayed as available drive letters” on page 44 v “Notification email for restore session contains an IPv6 address” on page 44 v “Restore session fails when using vFiler root volume” on page 44 v “Unable to retrieve deleted backup in offline session of Restore Agent” on page 45 36 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes v “Unable to download Restore Agent setup file using Internet Explorer on Windows 2003” on page 45 Error occurs during SnapManager for Microsoft SQL backup operation on Microsoft SQL databases An error occurs during a SnapManager for Microsoft SQL backup operation on Microsoft SQL databases stored on virtual disks (VMDKs) or during an unmount operation on a backup copy using Backup and Recovery. The error occurs in both instances when unmounting a datastore mounted on an ESX host that is contained in an ESX HA cluster. The datastore remains mounted on the ESX host, along with the underlying cloned storage on the storage system. This results in the following error message: The vSphere HA agent on host failed to quiesce file activity on datastore Workaround: You can either remove the datastore from the ESX host or remove the cloned volume or LUN from the storage system. 1. Unmount the datastore using the vSphere client by finding the datastore name using the following format: (Backup vmdk_). Note: If the original datastore name is sql_db and the name of the backup copy is sqlsnap__SMSQLVM_03-11-2013_16.14.04, then the name of the datastore would be sql_db (Backup vmdk_sqlsnap_smsql_03-132013_16.14.04). 2. Remove the cloned volume or LUN from the storage system by finding the cloned volume using the following format: . Note: If the name of the original volume is sql_db_vol, then the name of the cloned volume would be sql_db_vol_mount_vmdk_. You must take offline the cloned volume or LUN before you can destroy it. 3. If you are using a VMFS datastore, you must rescan the appropriate iSCSI or FCP initiator to completely remove the VMFS datastore. Backup and Recovery fails to enumerate the NFS exports when using an alias IP address Backup and Recovery does not list the storage entities when you access the storage system using an alias IP address, such as when an NFS volume is exported to the vCenter Server in Virtual Storage Console 4.1 for VMware vSphere. Known issues 37 When the storage system is not a vFiler or vFiler0 unit and the interfaces are exported to the vCenter Server using an alias IP address, the datastores that are exported using the alias IP address are not found by Backup and Recovery. This issue was found in Data ONTAP 8.0.1 and might occur only on certain versions of Data ONTAP. This results in the following error: Unable to manage datastore datastore-xxxx: Failed to find NFS export path /vol/xxxx for datastore xxxxxx on storage system xxxxx. NFS datastores must be exported from a Data ONTAP storage system. Workaround: If you need to access the storage system using an alias IP address, you must create the volumes on a vFiler unit. Insufficient privileges error occurs and Backup and Recovery incorrectly reports the error Attempting to create a backup job fails on a storage system for which a user does not have sufficient Data ONTAP privileges. This error is not reported correctly in the Backup and Recovery GUI. The GUI displays an EAPIERROR error instead of the correct error message. There is no other indication in the GUI that indicates a failure because of insufficient privileges for a storage system. Workaround: In the VSC logs, this error message contains details that indicate privilege issues. You can refer to the VSC logs for more details at VSC_Installation_Directory/logs/smvi.log. Attempting to create an on-demand backup job to back up only the virtual machine fails with an error In Backup and Recovery, when you specify a one-time backup job to back up only the virtual machine and not the entire datastore, the backup job fails with an error. Workaround: You must change your SMVI service log from a local system account to This Account and then you can set up your system with the vCenter administrator credentials. Discovery of FC and iSCSI datastores might fail if the FC or iSCSI feature is not enabled on some hardware platforms The Backup and Recovery capability might fail to discover FC or iSCSI datastores on certain systems unless the licensed_feature.fcp.enable or licensed_feature.iscsi.enable options are enabled on those systems. The Backup and Recovery capability does not check the state of the option and operates as if the FC or iSCSI license is enabled on the system. 38 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Workaround: You can check if the FC or iSCSI license is enabled by using the license command and then enabling the corresponding licensed_feature.fcp.enable or licensed_feature.iscsi.enable option on your system. Restoring a VMFS virtual machine causes loss of independent disks If you restore a virtual machine that resides on a VMFS datastore and has an independent disk attached (either persistent or nonpersistent), the independent disks, when excluded from a backup, cannot be restored. Workaround: Save these independent disks in their respective datastores after the restore operation finishes by detaching them and reattaching them as follows: 1. Click Edit VM Settings. 2. Remove the independent disk. 3. Restore the virtual machine from the backup. 4. Click Edit VM Settings and reattach the independent disk. VMDKs should be removed from original datastore After you restore a VMFS virtual machine, VMDKs should be removed from the original datastore and appear in the newly created folder on the destination datastore. Typically, the folder name consists of the original name with a sequence number added to it. Workaround: If the VMDK remains in the original datastore after restoring the virtual machine, you might want to delete it, to free space on the datastore. Before you delete any files or folders on the original datastore, review the contents of the destination folder to ensure that the backup restored properly. List of available virtual entities does not change if two datacenters have the same name In the Backup wizard, the list of available virtual entities that you can select for a backup job does not change depending on the datacenter that you select if two datacenters have the same name. Typically, the virtual entities corresponding to the selected datacenter should appear. Workaround: When you create a backup job, you must select a datacenter in the Inventory tree and then select the specific virtual machine or datastore corresponding to that datacenter. Restoring a VMDK to an alternate datastore causes loss of the original VMDK on a virtual machine If there is not enough space on the alternate datastore to hold the VMDK, the restore operation fails and you receive an error message stating that resources are unusable on the virtual machine. Known issues 39 Workaround: When restoring a VMDK, ensure that there is enough free space on the datastore for the VMDK. Virtual machine cannot power on after NFS or VMFS datastore is restored You cannot power on the virtual machine after restoring the datastore if the underlying volume has run out of space during the restore operation. Workaround: Provision more space on the volume and try to manually power on the virtual machine. Unable to delete restore session even if source virtual machine has been deleted Using the force option fails to delete a restore session even if the source virtual machine has been deleted. The restore session is automatically removed when the session expires and no user action is necessary Successful backup job might trigger a notification during backup When a scheduled backup job is configured to send email for warnings or errors, a successful backup might trigger a warning, because the Backup and Recovery capability was unable to create any vCenter tasks during the backup. Even though the email you receive from the Backup and Recovery capability indicates a warning, the backup job completes successfully. Workaround: Re-register VSC for VMware vSphere with the vCenter Server or restart the vCenter Server. Virtual machine is unresponsive during a backup When you back up a datastore of a powered-on virtual machine that is experiencing heavy I/O activity, the vCenter Server might time out. During this time, the host might be disconnected from the vCenter Server. Workaround: Check the performance of your storage system and make sure you have installed the latest VMware patches. Running a script during backup results in an error Running a script using perl script.pl in the Backup and Recovery capability user interface or CLI does not work for the Perl system command. Workaround: You must remove the word Perl from the system command and specify script.pl in the script. Backup and Recovery cannot verify whether the scripts ran successfully If the path for Perl was not set up correctly, the Backup and Recovery capability might not be able to verify whether the Perl script ran successfully during pre-backup or post-backup phase. 40 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes vCenter Server displays an error message that virtual machine has two VMDKs with same UUID When using the Restore Agent or the Backup and Recovery user interface to perform an offline disk attachment, you might receive an error message about duplicate Universally Unique Identifiers (UUIDs). Workaround: To allow duplicate UUIDs, complete the following steps: 1. Use Restore Agent to mount a disk from a backup copy of the virtual machine. 2. After mounting the disk, launch the vSphere Client. 3. Select a virtual machine in the vSphere Client window. 4. Ensure that the VMware administrator does not need to reply to any questions in the vSphere Client window. 5. You might need to click the first two tabs in the vSphere Client for the virtual machine. 6. Attach a disk from a backup copy of the virtual machine, using either the Backup and Recovery user interface or the VSC CLI. 7. Repeat Steps 2 through 6. Automount is enabled by default The automount feature is enabled by default on Windows (XP, Vista, and Windows 7) operating systems, and the feature cannot be disabled. Windows automatically assigns a drive letter and mounts the disks. In offline mode, disks are mounted to a drive letter when they are attached to the virtual machine. Unable to verify which disk is being restored During backup and restore of a virtual disk, the disk is identified by the name used in vCenter Server (such as Hard Disk 1). There is no correlation in SnapManager for Virtual Infrastructure between this name and the drive letter used by the underlying virtual machine operating system. Workaround: Before running a restore operation, mount the backup and manually verify that the disk corresponds to the drive to be restored. Virtual machine is unresponsive during a backup that uses a VMware snapshot Running a backup job that uses a VMware snapshot might result in the loss of remote desktop connectivity and the virtual machine being unresponsive until the backup is finished running. Workaround: Run the backup job without using a VMware snapshot. Known issues 41 CLI times out during a backup During a backup, if the SnapManager for Virtual Infrastructure server does not respond in a timely manner, the CLI times out and the following message appears: Connection to SMVI server was refused. Ensure that the server is running. Workaround: If the Backup and Recovery capability detects this error, make sure that the server is running and then perform the backup. Unable to mount an NFS volume if the export policy for the parent volume and the mounted backup volume are not the same When you are mounting a backup, if the parent NFS volume has a customized export policy but the mounted backup volume has a default export policy, an error occurs and the mount operation fails. Workaround: If you are mounting a backup copy to a different ESX server than the server on which the parent NFS volume is mounted, you must make sure that the export policy for the parent volume is the same as the export policy for the mounted backup volume. If you are using a customized export policy for the parent volume, then you must mount the backup to the same ESX server on which the volume was initially mounted and not to a different ESX server. Disconnecting the FC initiator used to mount a LUN clone from the ESX server results in the mounted backup being removed from the server The Backup and Recovery capability does not allow you to specify which initiators (iSCSI or FC) you want to have a LUN mapped to. If you back up a datastore of an iSCSI-based LUN and then mount it on the host, the Backup and Recovery capability might use FC instead of iSCSI initiators. If you then remove the FC initiator from the storage system, the mounted backup is removed from the server. Mounted backup not automatically removed when the restore session expires Although the restore session and mounted backup should be automatically removed when the restore session expires, this did not occur. Workaround: You must restart the SnapManager for Virtual Infrastructure server service and delete the SFR session manually, as follows: 1. From All Programs > Administrative Tools > Services, right-click SnapManager VI service in the list of services and select Start. 2. From Home > Solutions and Applications, select IBM > Backup and Recovery > Single File Restore, select the expired restore session ID, and click Delete. 42 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Single file restore on a Windows Server 2008 sometimes fails to bring disk online When you use the single file restore feature of SnapManager for Virtual Infrastructure to mount a VMDK to the same Windows Server 2008 R2 virtual machine from which the backup was taken, it sometimes fails to bring the disk online. In the Windows Server 2008 R2 release, the SAN policy was changed to prevent disks from automatically coming online if the disk signature matches an existing disk signature. If you have a Limited Self Service Restore job, you must manually mark the disk online using a tool such as diskpart or the Disk Management MMC plug-in. After you do that, drive letters are assigned to the partitions. If you have a Self Service restore job, you are prompted to assign a drive letter. In this case, single file restore then automatically brings the disk online and assigns the drive letter to the first partition on the disk. If this is the system-reserved partition, it is the only partition that receives a drive letter. If you have more partitions on the disk, you must manually assign the drive letter using a tool such as diskpart or the Disk Management MMC plug-in. Single file restore does not automatically support VMDKs with multiple partitions The single file restore feature of SnapManager for Virtual Infrastructure automatically assigns a drive letter to a single partition on a restored VMDK. If there are other partitions on the VMDK, you must manually assign the drive number or mount point. This issue occurs because Backup and Recovery Restore Agent only sees the first drive letter assigned to a VMDK. As a result, it only presents the option to mount the drive letter of the first partition. You can manually assign drive letters using a tool such as diskpart or the Disk Management MMC plug-in. After you manually add drive letters for the additional partitions, the Backup and Recovery Restore Agent recognizes them, so you do not need to perform any manual steps to remove the drive letters. Attaching a .sfr file using Outlook Web Access (OWA) might fail A restore session configuration might not be downloaded correctly if it is attached as a .sfr file using Outlook Web Access (OWA). Workaround: Check the Microsoft support site for more information about OWA server configuration or access the email message using a non-OWA mailbox. Known issues 43 Deleting a nonexistent restore session results in an error When you attempt to delete a nonexistent restore session using the name of a valid virtual machine or a virtual machine for which this session does not exist, an error message appears in the CLI output. Drive letters that are used for network mapped drives might be incorrectly displayed as available drive letters Drive letters that are used for network mapped drives might be displayed as available drive letters in the Restore Agent of the Backup and Recovery capability because of a Microsoft API bug. Notification email for restore session contains an IPv6 address When you create a restore session in Backup and Recovery, the notification email for the restore session contains an IPv6 address for the virtual machine instead of an IPv4 address. Workaround: You can prevent IPv6 addresses from appearing in the notification email. You only need to complete this procedure once and the system generates the correct IPv4 addresses. To avoid IPv6 addresses appearing in the notification email for a restore session, complete the following steps: 1. Shut down the SnapManager for Virtual Infrastructure server (by using Windows Services). 2. Open the wrapper.conf file. This file is stored in the following location: Virtual Storage Console\smvi\server\etc\wrapper.conf 3. Locate the following section of the file: # Java Additional Parameters wrapper.java.additional.1=-XX:MaxPermSize=128m wrapper.java.additional.2=-Dcom.sun.management.jmxremote wrapper.java.additional.3=-Dcommon.dir=. wrapper.java.additional.4=Dorg.apache.cxf.Logger=org.apache.cxf.common.logging.Log4jLogger 4. Add the following code to this section of the file: wrapper.java.additional.5=-Djava.net.preferIPv4Stack=true 5. Start the SnapManager for Virtual Infrastructure server (by using Windows Services) and verify the email notification for the restore session. Restore session fails when using vFiler root volume The restore operation for a virtual machine using the vFiler root volume results in the following error: 44 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Storage system error: The vfiler root volume vol4041vf1 cannot have clones. Workaround: Remove the FlexClone license or set storage.provision.volumeCloneLun=false in smvi.override and restart the SnapManager for Virtual Infrastructure server. Unable to retrieve deleted backup in offline session of Restore Agent You cannot retrieve deleted data from a backup mounted in an offline session of Restore Agent. For a limited self-service restore session, any changes that you make to the disks in a backup remain until the restore session ends, because the volume clone is removed only after the session expires and the disks are removed. Workaround: You need to re-create or generate a new restore session and mount the backup again so that you can retrieve the deleted backup. Unable to download Restore Agent setup file using Internet Explorer on Windows 2003 Using Internet Explorer on Windows 2003 to download the Restore Agent setup file might result in a message saying that Internet Explorer cannot download the file. Workaround: You must modify the security settings for the Internet Explorer browser. For details, see the documentation for your version of the browser. Known issues with the application programming interface for vCloud Director It is important for you to be familiar with issues applying to the application programming interface (API) for vCloud Director. v “vApp operation defaults to vCloud Director operation if vNIC connected to organization vDC network” v “Invalid error messages seen when cloning a vAPP with an NFS datastore ” on page 46 vApp operation defaults to vCloud Director operation if vNIC connected to organization vDC network When performing a cloning or provisioning operation for a vApp that contains virtual machines using a vNIC connected to an organization vDC network, the operation defaults to a vCloud Director operation. vCloud Director operations require more time and resources because they do not use IBM N Series file clone feature, which uses fast, space-efficient clones to perform these operations. Known issues 45 To prevent this issue from occurring, make sure any vNICs you use are not connected to an organization vDC network. Invalid error messages seen when cloning a vAPP with an NFS datastore When performing a cloning operation on a vAPP that contains an NFS datastore, Virtual Storage Console for VMware vSphere writes invalid error messages to the kamino.log. These invalid errors have been seen on storage systems running Data ONTAP operating in 7-Mode. The following invalid error messages have been seen: NaException netapp.manage.NaAPIFailedException: File exists (errno=17) NaException netapp.manage.NaAPIFailedException: No such file or directory (errno=2) Workaround: There is no workaround. If you encounter one of these error messages when using VSC to clone virtual machines on an NFS datastore, you can safely ignore the error messages. 46 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Fixed issues VSC for VMware vSphere includes product fixes that you should know about. In many cases, there is a workaround if an issue cannot be resolved. The following sections contain information about these issues. The issues are organized according to the VSC for VMware vSphere capability to which they apply. v “Fixed issues: Monitoring and Host Configuration” v “Fixed issues: Provisioning and Cloning” v “Fixed issues: Optimization and Migration” on page 48 v “Fixed issues: Backup and Recovery” on page 49 Fixed issues: Monitoring and Host Configuration Some known issues are fixed in the Monitoring and Host Configuration capability of Virtual Storage Console 4.2 for VMware vSphere. IndexOutOfBoundsException no longer occurs when storage system In previous versions of Virtual Storage Console for VMware vSphere, Monitoring and Host Configuration sometimes triggered an IndexOutOfBoundsException when it updated information about a storage system, resulting in a vCenter Server task reporting an error state. This issue was only seen when you used VSC with 9 or more IP addresses. VSC 4.2 resolves this issue. Discovery and update tasks now complete correctly In previous versions of Virtual Storage Console for VMware vSphere, the discovery and update processes sometimes appeared to hang and not complete when using vCenter Server 5.1. VSC 4.2 resolves this issue. Fixed issues: Provisioning and Cloning Some known issues are fixed in the Provisioning and Cloning capability of Virtual Storage Console 4.2 for VMware vSphere. © Copyright IBM Corp. 2013 47 Issue where ESX hosts lost access to NFS datastore when it was mounted on another host resolved In versions 4.0 and 4.1 of Virtual Storage Console for VMware vSphere, the mount feature of Provisioning and Cloning overrode the existing settings. As a result, if the NFS datastore was currently mounted on one or more ESX hosts, then mounting the datastore on another ESX hosts could break the connection to the other ESX hosts. When the connection broke, the datastore still appeared in the inventories of those ESX hosts, but the hosts could not access any data on it. This issue has been fixed in VSC 4.2. Provisioning to vFiler units if interfaces were selected in Resources dialog now succeeds Prior to version 4.2 of Virtual Storage Console for VMware vSphere, operations would fail if you used the Provisioning and Cloning Resources dialog to select the interface. This issue has been fixed in VSC 4.2. Resources dialog adds support for multiple addresses for an interface Starting in version 4.2 of Virtual Storage Console for VMware vSphere, the Resources dialog correctly handles multiple addresses, such as addresses using aliases, for an interface. Fixed issues: Optimization and Migration Some known issues that affected Optimization and Migration have been fixed in this release of VSC for VMware vSphere. Issue with Virtual Machine Alignment panel loading slowly resolved With versions Virtual Storage Console for VMware vSphere prior to 4.2, the provided Virtual Machine Alignment panel could take from 30 seconds to several minutes to load in environments with a large number of virtual machines. In some cases, the panel failed to load. This issue has been fixed in VSC 4.2. 48 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Fixed issues: Backup and Recovery Some known issues that affected Backup and Recovery have been fixed in this release of Virtual Storage Console for VMware vSphere. Mounted backup now removed when restore session expires In previous versions of Virtual Storage Console for VMware vSphere, the restore session and mounted backup were not automatically removed when the restore session expired. This issue has been resolved in VSC 4.2. Backup and Recovery now accepts email addresses containing hyphens In version 4.1 of Virtual Storage Console for VMware vSphere, Backup and Recovery displayed an error message stating that an email address was invalid if it contained hyphens. This issue has been fixed in VSC 4.2. Timeout issue with SMSQL backups resolved Version 4.2 of Virtual Storage Console for VMware vSphere resolves a problem where SMSQL backups sporadically failed to complete because of a timeout issue. Single file restore session now removed at scheduled time In previous versions of Virtual Storage Console for VMware vSphere, Backup and Recovery sometimes failed to remove a single file restore session at the scheduled time. This issue has been fixed in VSC 4.2. "$" character in a LUN serial number now accepted Version 4.2 of Virtual Storage Console for VMware vSphere fixes an issue that prevented Backup and Recovery from accepting the "$" character in a LUN serial number. In earlier versions of VSC, Backup and Recovery displayed the error message Failed to Reconfig instead of connecting a LUN from SnapDrive for UNIX or Windows if the LUN serial number contained a "$" character. Fixed issues 49 50 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Where to find additional information The following table describes documents with the most current information about host requirements, supported configurations, and procedures for installing and managing Virtual Storage Console 4.1 for VMware vSphere. To learn about... Go to this information source... New and changed features, limitations, known Virtual Storage Console for VMware vSphere Release issues, fixed issues, and documentation corrections Notes How to download and install VSC for VMware vSphere Virtual Storage Console for VMware vSphere Installation and Administration Guide How to discover and add storage controllers in VSC for VMware vSphere Virtual Storage Console for VMware vSphere Installation and Administration Guide How to collect VSC for VMware vSphere log files Virtual Storage Console for VMware vSphere Installation and Administration Guide How to provision or clone a virtual machine Virtual Storage Console for VMware vSphere Installation and Administration Guide How to perform online alignment and migration of virtual machines Virtual Storage Console for VMware vSphere Installation and Administration Guide How to add storage systems using the Backup and Virtual Storage Console for VMware vSphere Recovery capability Installation and Administration Guide How to back up a virtual machine or datastore using the Backup and Recovery capability Virtual Storage Console for VMware vSphere Installation and Administration Guide How to find troubleshooting information Virtual Storage Console for VMware vSphere Installation and Administration Guide How to find preference files, logs, and export files Virtual Storage Console for VMware vSphere Installation and Administration Guide How to find programmable APIs Virtual Storage Console for VMware vSphere Installation and Administration Guide The latest supported configurations www.ibm.com/storage/support/nseries/ Supported Fibre Channel SAN topologies The FC and iSCSI Configuration Guide VSC for VMware vSphere GUI VSC for VMware vSphere online Help Backup and Recovery GUI Backup and Recovery online Help © Copyright IBM Corp. 2013 51 Configuring your host for the Host Utilities The HP-UX Host Utilities Installation and Setup Guide The Linux Host Utilities Installation and Setup Guide The AIX Host Utilities Installation and Setup Guide The Windows Host Utilities Installation and Setup Guide 52 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Websites IBM maintains pages on the World Wide Web where you can get the latest technical information and download device drivers and updates. The following web pages provide N series information: v A listing of currently available N series products and features can be found at the following web page: www.ibm.com/storage/nas v The IBM System Storage N series support website requires users to register in order to obtain access to N series support content on the web. To understand how the N series support web content is organized and navigated, and to access the N series support website, refer to the following publicly accessible web page: www.ibm.com/storage/support/nseries/ This web page also provides links to AutoSupport information as well as other important N series product resources. v IBM System Storage N series products attach to a variety of servers and operating systems. To determine the latest supported attachments, go to the IBM N series interoperability matrix at the following web page: www.ibm.com/systems/storage/network/interophome.html v For the latest N series hardware product documentation, including planning, installation and setup, and hardware monitoring, service and diagnostics, see the IBM N series Information Center at the following web page: http://publib.boulder.ibm.com/infocenter/nasinfo/nseries/index.jsp v See the Independent software vendors (ISV) Solutions Resource Library and VMware content at the following web page: www-03.ibm.com/systems/storage/solutions/isv/isv_vmware.html © Copyright IBM Corp. 2013 53 54 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Copyright and trademark information This section includes copyright and trademark information, and important notices. Copyright information Copyright ©1994 - 2013 Network Appliance, Inc. All rights reserved. Printed in the U.S.A. Portions copyright © 2013 IBM Corporation. All rights reserved. US Government Users Restricted Rights - Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp. No part of this document covered by copyright may be reproduced in any form or by any means - graphic, electronic, or mechanical, including photocopying, recording, taping, or storage in an electronic retrieval system—without prior written permission of the copyright owner. References in this documentation to IBM products, programs, or services do not imply that IBM intends to make these available in all countries in which IBM operates. Any reference to an IBM product, program, or service is not intended to state or imply that only IBM’s product, program, or service may be used. Any functionally equivalent product, program, or service that does not infringe any of IBM’s or NetApp’s intellectual property rights may be used instead of the IBM or NetApp product, program, or service. Evaluation and verification of operation in conjunction with other products, except those expressly designated by IBM and NetApp, are the user’s responsibility. No part of this document covered by copyright may be reproduced in any form or by any means - graphic, electronic, or mechanical, including photocopying, recording, taping, or storage in an electronic retrieval system—without prior written permission of the copyright owner. Software derived from copyrighted NetApp material is subject to the following license and disclaimer: THIS SOFTWARE IS PROVIDED BY NETAPP "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, WHICH ARE HEREBY DISCLAIMED. IN NO EVENT SHALL NETAPP BE LIABLE FOR ANY DIRECT, INDIRECT, © Copyright IBM Corp. 2013 55 INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. NetApp reserves the right to change any products described herein at any time, and without notice. NetApp assumes no responsibility or liability arising from the use of products described herein, except as expressly agreed to in writing by NetApp. The use or purchase of this product does not convey a license under any patent rights, trademark rights, or any other intellectual property rights of NetApp. The product described in this manual may be protected by one or more U.S.A. patents, foreign patents, or pending applications. RESTRICTED RIGHTS LEGEND: Use, duplication, or disclosure by the government is subject to restrictions as set forth in subparagraph (c)(1)(ii) of the Rights in Technical Data and Computer Software clause at DFARS 252.277-7103 (October 1988) and FAR 52-227-19 (June 1987). Trademark information IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. A complete and current list of other IBM trademarks is available on the Web at http://www.ibm.com/legal/copytrade.shtml Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both. Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both. UNIX is a registered trademark of The Open Group in the United States and other countries. NetApp, the NetApp logo, Network Appliance, the Network Appliance logo, Akorri, ApplianceWatch, ASUP, AutoSupport, BalancePoint, BalancePoint Predictor, Bycast, Campaign Express, ComplianceClock, Cryptainer, CryptoShred, Data ONTAP, DataFabric, DataFort, Decru, Decru DataFort, DenseStak, Engenio, Engenio logo, E-Stack, FAServer, FastStak, FilerView, 56 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes FlexCache, FlexClone, FlexPod, FlexScale, FlexShare, FlexSuite, FlexVol, FPolicy, GetSuccessful, gFiler, Go further, faster, Imagine Virtually Anything, Lifetime Key Management, LockVault, Manage ONTAP, MetroCluster, MultiStore, NearStore, NetCache, NOW (NetApp on the Web), Onaro, OnCommand, ONTAPI, OpenKey, PerformanceStak, RAID-DP, ReplicatorX, SANscreen, SANshare, SANtricity, SecureAdmin, SecureShare, Select, Service Builder, Shadow Tape, Simplicity, Simulate ONTAP, SnapCopy, SnapDirector, SnapDrive, SnapFilter, SnapLock, SnapManager, SnapMigrator, SnapMirror, SnapMover, SnapProtect, SnapRestore, Snapshot, SnapSuite, SnapValidator, SnapVault, StorageGRID, StoreVault, the StoreVault logo, SyncMirror, Tech OnTap, The evolution of storage, Topio, vFiler, VFM, Virtual File Manager, VPolicy, WAFL, Web Filer, and XBB are trademarks or registered trademarks of NetApp, Inc. in the United States, other countries, or both. All other brands or products are trademarks or registered trademarks of their respective holders and should be treated as such. NetApp, Inc. is a licensee of the CompactFlash and CF Logo trademarks. NetApp, Inc. NetCache is certified RealSystem compatible. Copyright and trademark information 57 58 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Notices This information was developed for products and services offered in the U.S.A. IBM may not offer the products, services, or features discussed in this document in other countries. Consult your local IBM representative for information on the products and services currently available in your area. Any reference to an IBM product, program, or service is not intended to state or imply that only that IBM product, program, or service may be used. Any functionally equivalent product, program, or service that does not infringe on any IBM intellectual property right may be used instead. However, it is the user's responsibility to evaluate and verify the operation of any non-IBM product, program, or service. IBM may have patents or pending patent applications covering subject matter described in this document. The furnishing of this document does not give you any license to these patents. You can send license inquiries, in writing to: IBM Director of Licensing IBM Corporation North Castle Drive Armonk, N.Y. 10504-1785 U.S.A. For additional information, visit the web at: http://www.ibm.com/ibm/licensing/contact/ The following paragraph does not apply to the United Kingdom or any other country where such provisions are inconsistent with local law: INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION “AS IS” WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer of express or implied warranties in certain transactions, therefore, this statement may not apply to you. This information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein; these changes will be incorporated in new editions of the publication. IBM may make © Copyright IBM Corp. 2013 59 improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time without notice. Any references in this information to non-IBM web sites are provided for convenience only and do not in any manner serve as an endorsement of those web sites. The materials at those web sites are not part of the materials for this IBM product and use of those web sites is at your own risk. IBM may use or distribute any of the information you supply in any way it believes appropriate without incurring any obligation to you. Any performance data contained herein was determined in a controlled environment. Therefore, the results obtained in other operating environments may vary significantly. Some measurements may have been made on development-level systems and there is no guarantee that these measurements will be the same on generally available systems. Furthermore, some measurement may have been estimated through extrapolation. Actual results may vary. Users of this document should verify the applicable data for their specific environment. Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. If you are viewing this information in softcopy, the photographs and color illustrations may not appear. 60 IBM System Storage N series: Virtual Storage Console 4.2 for VMware vSphere Release Notes Notices 61
 NA 210-05938_A0, Printed in USA GC52-1351-10