IBM Support

Known Issues and Limitations: Version 4.1.x IBM Spectrum Protect™ Snapshot for VMware

Preventive Service Planning


Abstract

This document details the known issues and limitations of the IBM Spectrum Protect™ Snapshot for VMware V4.1.x component.

Content

Tip: Beginning with Version 4.1.3, IBM Tivoli Storage FlashCopy Manager is now IBM Spectrum Protect™ Snapshot. Some applications such as the software fulfillment systems and IBM License Metric Tool use the new product name. However, the software and its product documentation continue to use the Tivoli Storage Manager product name. To learn more about the rebranding transition, see http://www.ibm.com/support/docview.wss?uid=swg21963634.


Links to the Known Issues and Limitations information for the other components that are a part of IBM Spectrum Protect™ Snapshot V4.1.x are located at the following URL: http://www.ibm.com/support/docview.wss?uid=swg21644477


This document is divided into linked sections for ease of navigation. You may use the links below to jump to the desired section of the document:

Known Issues and Limitations


Resolved Issues with FlashCopy Manager 4.1.1.0
Issues and Limitations specific to FlashCopy Manager 4.1.1.0 and later modifications
Resolved Issues with FlashCopy Manager 4.1.2.0
Issues and Limitations specific to Flashcopy Manager 4.1.2.0 and later modifications
Resolved Issues with IBM Spectrum Protect™ Snapshot for VMware 4.1.3.0
Issues and Limitations specific to IBM Spectrum Protect™ Snapshot for VMware 4.1.3.0 and later modifications
Resolved Issues with IBM Spectrum Protect™ Snapshot for VMware 4.1.4.0
Issues and Limitations specific to IBM Spectrum Protect™ Snapshot for VMware 4.1.4.0 and later modifications



Known Issues and Limitations



Known Issues and Limitations Specific to the IBM Spectrum Protect™ Snapshot for VMware Installation and Configuration

  • SELinux must be disabled

    SELinux is activated by default on Red Hat Enterprise Linux. However, SELinux should be temporarily disabled when installing or uninstalling IBM Spectrum Protect™ Snapshot.

    To temporarily disable SELinux before uninstall, enter the following:

    setenforce 0

    To re-enable SELinux after completing the installation or uninstall process, enter the following:

    setenforce 1

  • Insufficient space on root file system

    If the root file system is running out of space during IBM Spectrum Protect™ Snapshot installation and/or setup, the process might hang. Always ensure there are sufficient space reserves in the root file system.

  • Installation in console mode requires X Window server

    In combination with some ssh clients, the console-based installation incorrectly checks for the presence of a X Window server. Therefore, if a X Window server is not available during IBM Spectrum Protect™ Snapshot installation you may see this error:

    Launching installer...

    X connection to localhost:10.0 broken (explicit kill or server shutdown).

    User response: Unset the DISPLAY environment variable (unset DISPLAY) and re-run the installer. (77219)

  • Installation when GSKit v8.0.13 is present.

    Installing IBM Spectrum Protect™ Snapshot v4.1.x on a Linux machine with GSKit packages of version "8.0.13" already installed will fail with the following error: "Fatal error: No appropriate GSKit installation found."
    rpm -qa | grep gsk will show:
    gskcrypt64-8.0-14.14
    gskssl64-8.0-14.14
    This indicates that the GSKit packages were successfully upgraded to the version delivered with IBM Spectrum Protect™ Snapshot. However, due to a flaw in the post-install scripts of GSKit 8.0.13, PATH variables are not set correctly so that the GSKit command line tools are not found. The issue was fixed with GSKit version 8.0.13.1.

    Work-around: Before installing IBM Spectrum Protect™ Snapshot, uninstall GSKit version 8.0.13. When IBM Spectrum Protect™ Snapshot is installed it will also install GSkit version 8.0.14.14. (81744)

  • "Previous" button of uninstallation agent

    Using the "Previous" and "Next" buttons in the uninstallation wizard after the package has already been uninstalled may lead to an exception. (78014)

  • When uninstalling in console mode , the uninstaller may report "Some items could not be removed" although uninstallation completed successfully. (77924)

  • During IBM Spectrum Protect™ Snapshot installation you may see the title of some installer panels being truncated in languages other than English. (74278, 76087)

  • During IBM Spectrum Protect™ Snapshot installation you may see that the installer offers you the selection of a quit action but when you select the quit option you receive the following error:
    "This step does not allow quitting the installer at this moment." (74891)
  • In silent installation mode, the installation finishes without errors or warnings even when specifying a wrong value for either hostname, username, or password in the response file. Accordingly, take care of entering the correct values. (75725)
  • During the installation of IBM Spectrum Protect™ Snapshot for VMware in advanced mode, if a user other than "tdpvmware" is entered, which is an existing user, it will not be added to the tdpvmware group. As a consequence, IBM Spectrum Protect™ Snapshot is not operational until the user is added to the group. (90257)

    Work-around: Right after the installation is completed, add the specified user manually to the tdpvmware group, for example, with the usermod command. Ensure that you start a fresh shell session afterward so that the change becomes effective.


Known Issues and Limitations Specific to the IBM Spectrum Protect™ Snapshot for VMware GUI

  • When using the vSphere client GUI and launching the plug-in GUI of IBM Spectrum Protect™ Snapshot for VMware, the plug-in is loaded into an instance of the Internet Explorer web browser installed on the workstation. It has been observed that over time, as the plug-in is used, the overall memory consumption by the vSphere client process can increase and remain at the increased level, reflecting increased memory consumption by the Internet Explorer browser. This issue is especially noticeable when using Internet Explorer version 8.

    This same increased memory problem can also occur when launching the plug-in from the vSphere 5.1 web client using a stand-alone browser, and again the problem is most significant with Internet Explorer version 8.

    Depending on duration of usage and on the amount of memory available on the workstation hosting the application, this problem may impact performance of applications running on the workstation.

    Work-around: Close the vSphere client application or stand-alone browser. This will release all memory allocated by the browser. Log back into the vSphere client and launch the plug-in to resume use. (86646)

  • The titles of some pop-up windows in the plug-in GUI may be truncated in languages other than English. (74279)

  • Datastore restore wizard provides registration checkbox for ineligible VMs

    When using the restore wizard to restore datastores, you can use the restore wizard to select which VMs to register at the completion of the restore process. VMs that are not eligible for registration are identified with a yellow warning icon. However, the restore wizard allows you to select VMs that are not eligible for registration. Even though these VMs are selected, they still will not be registered. This is a known problem.

    Workaround
    : To resolve this issue, ignore the selection box for VMs that are not eligible for registration. View the Registration Exceptions page in the restore wizard to see the complete list of VMs that will not be registered. (99029)
    This limitation is lifted with v4.1.1.0.
  • Attach VM dialog does not display second ESX host when first ESX host contains large number of VMs

  • In this scenario, a backup task is selected for restore using the IBM Spectrum Protect™ Snapshot for VMware GUI. In the Attach VM dialog, an ESX host is selected and expanded to view the VM backups that are available to attach. When a second ESX host is selected and expanded, the VM backups on the second ESX host do not display. This display issue occurs when the first expanded ESX host contains a large number of VMs. This is a known issue.

    Workaround: To resolve this issue, complete one of the following tasks:
    - When this scenario occurs, collapse the first ESX host, then expand the second ESX host.
    - When the ESX host where the VM resides is already known, only expand that ESX host. (98775)
  • When creating an IBM Spectrum Protect™ Snapshot for VMware backup schedule with offload to IBM Spectrum Protect™ , it is mandatory to select a data mover in the "Destination" panel of the "Create Schedule" wizard. When editing the properties of an existing schedule there is no drop down box to change the data mover on the "Destination" panel.

    Workaround:
     Creating a new schedule with a different data mover selected. (120997)
  • When restoring an an IBM Spectrum Protect™ Snapshot for VMware backup that contains a large number of VMs (app. 500) the instant restore wizard stop working on the VM selection panel.

    Workaround:
     None. This problem will be fixed with an APAR. (120717)



Known Issues and Limitations Specific to the IBM Spectrum Protect™ Snapshot for VMware Component (except GUI)

  • Limitation: Virtual machines that have disks relying in different directories of the same datastore are not supported by IBM Spectrum Protect™ Snapshot for VMware. This applies especially to OpenStack environments.

    For example the virtual machine specified by the file nova_vm01.vmx is not supported if it has the following structure:
[datastore1] nova_vm01/nova_vm01.vmx
[datastore1] nova_vm01/nova_vm01.vmdk
[datastore1] cinder_vm01_1/cinder_vm01_1.vmdk
  • The attempt to backup a VMFS datastore residing on SAN storage with a DEVICE_CLASS configured with COPYSERVICES_HARDWARE_TYPE = NAS_NSERIES results in the following error message (110170):

    ERROR FMM9202E Additional support information: An exception was thrown at position: esd-vsphereapiaccessor.cpp(2709).

    Work-around: Remove the VMFS datastore from the backup schedule or assign an appropriate DEVICE_CLASS configured for SAN storage. DEVICE_CLASS sections configured with COPYSERVICES_HARDWARE_TYPE = NAS_NSERIES can only be used for NFS datastores.
  • Limitation: Space Efficient Sparse Virtual Disks introduced with vSphere 5.1 are not supported by IBM Spectrum Protect™ Snapshot.
  • Limitation: Backup of a VM with virtual disks of another backup operation attached is not supported.

    IBM Spectrum Protect™ Snapshot for VMware 4.1.x does not support the backup of a VM with virtual disks from another backup attached to it.
    Depending on the storage subsystem, trying to backup such a VM can lead to various error messages, for example:

    #INFO ERROR during doFlash(), unable to add volume" TSM_A0GOZIXYO3_A0GRD3Q8DA_232E27EA_1_dup " to Consistency Group" TSM_A0GOZIXYO3_A0GRF1DMHG_GRF1E03X ".)

    (IC79638)

  • Limitation: Raw Device Mapping (RDM) disks of Virtual machines cannot be protected with IBM Spectrum Protect™ Snapshot. If a VM using RDM volumes is involved in a backup operation the RDM volumes will be implicitly excluded from the backup, but the backup of the other disks of this VM will be performed. When restoring such a backup, the RDM disks will not be restored, but only the other disks of the VM are restored.
  • Limitation: For Virtual machines with RDM disks, the backup modes SNAPSHOT_INCL_MEM and SNAPSHOT_EXCL_MEM are altered to mode ASIS. As a consequence, the backup of the VM is only crash consistent. Other virtual machines included in the same backup are not affected.

    Note: The backup mode change occurs for a IBM Spectrum Protect™ Snapshot for VMware backup only. The mode is not changed when you do a snapshot backup with a subsequent offload to Tivoli Storage Manager. The restore process from Tivoli Storage Manager server maintains a file system level consistency.

    See the IBM Tivoli Storage FlashCopy Manager for VMware v4.1 Installation and User's Guide, section "Raw device mapping" for more details.
  • Limitation: All datastores participating in a backup operation must reside on the same storage subsystem or SAN VC storage cluster.

  • Limitation: When using IBM System Storage DS8000 series only datastores with a single LUN are supported.

  • Limitation: Restores of distributed virtual machines spanning multiple datastores can be done to the original location only. This applies to both full VM and distinct virtual disk restores.

  • Limitation: Distributed virtual machines spanning multiple datastores are not supported on IBM System Storage DS8000 series.

  • Limitations for single-disk restore:

    For single-disk restore, virtual disks cannot be restored as additional disks to the original VM. However, restoring as original disk (potentially overwriting existing ones) is possible. In addition, they can be attached to the original VM as new disks; so file level restores can easily be done.

    Single disk restores to the original location are only possible if no VMware snapshots have been taken manually of these VMs before or after the backup.

    Given the case that a backup contains a VM X' resulting from a redirected restore of VM X it is not possible to restore a single virtual disk from VM X' to the original VM X or vice versa. As a consequence, VMs resulting from redirected restores are not appropriate for production purposes. Instead, cloned VMs should be created with vCenter tooling.

  • Limitation: The backup of a datastore containing VMs with duplicate UUIDs is not supported. In the case duplicate UUIDs are detected IBM Spectrum Protect™ Snapshot will abort the operation with an appropriate error message.
    Duplicate UUIDs may be caused by the following scenarios:
    1. Performing a redirected restore of a VM to the same or another datastore of this backup and then performing another backup of this datastore without starting up the restored VM prior this.
    2. Performing a redirected restore of a VM to the same or another datastore of this backup, then starting up the restored VM telling vCenter accidentally the VM has been moved (instead of copied) and finally performing another backup of this datastore again.

  • Virtual machines that use Fault Tolerance or VMDirectPath (hardware devices of the ESX/ESXi host that are passed through to a virtual machine) are not allowed to be suspended or to create snapshots by VMware. Thus IBM Spectrum Protect™ Snapshot will process such virtual machines always without suspending them or creating snapshots independent of the profile parameter setting for parameter VM_BACKUP_MODE and raise an according warning.

  • Backup and delete operations running concurrently

    If a FlashCopy backup operation is started while a delete operation with the force option is still running the backup operation may fail with message FMM4008E logged to the summary log file:

    DB 00:25:50 (712) FMM4008E File '/db2/S43/sqllib/acs/acsrepos/ori.1313619101000' cannot be opened. Reason: errno(2) No such file or directory DB 00:25:50 FMM9223E The operation will be aborted.

    Work-around: Make sure the delete operation completed by monitoring the log file for FlashCopy withdraw messages, then re-start the backup operation.
    Please note: A backup operation may take several minutes to complete. (IC79646)

  • SCSI controller persists after all virtual disks were detached.

    When virtual disks are attached to virtual machines SCSI controllers are needed. If the existing SCSI controller of the virtual machine has no free targets left IBM Spectrum Protect™ Snapshot for VMware creates a new SCSI controller for the virtual machine and attaches the virtual disks to this new SCSI controller. After detaching the disks this SCSI controller is left in the virtual machine configuration even if it does not have any disks attached to it.

    Work-around: Remove the obsolete SCSI controller manually from the VM using the vSphere client. (74287)

  • On full VM restore, all restored virtual disks turn into fully allocated disks even if they were thin-provisioned at the time of the backup. This might cause out-of-space issues. This does not affect the single disk restore and the instant restore operation.

    Work-around: If thin-provisioned disks are used make sure there is enough space in the target datastore(s) for the thick-provisioned size of all disks that are restored. (74894)

  • Corrupted vmcli database

    In rare cases, the vmcli database might get corrupted because the file service.properties is deleted. This can happen when the vmcli daemon process is killed while it is writing to this file. In this case, messages similar to the following example will appear in the vmcli log file:

    FUNCTIONS 15:03:27,346 (daemon) #ERROR FMM16119E An error occurred trying to create the Derby DB.
    com.ibm.storage.vmcli.exceptions.VmcliDBException: FMM16016E An error occurred while connecting to the Derby DB.
    (...)
    Caused by:
    java.sql.SQLException: DERBY SQL error: SQLCODE: -1, SQLSTATE: XJ041, SQLERRMC: Failed to create database 'VMCLIDB', see the next exception for details.::SQLSTATE: XBM0JDirectory /home/tdpvmware/tdpvmware/derby/ /home/tdpvmware/tdpvmware/derby/VMCLIDB already exists.
    (...)
    Caused by:
    org.apache.derby.client.am.SqlException: DERBY SQL error: SQLCODE: -1, SQLSTATE: XJ041, SQLERRMC: Failed to create database 'VMCLIDB', see the next exception for details.::SQLSTATE: XBM0JDirectory /home/tdpvmware/tdpvmware/derby/ /home/tdpvmware/tdpvmware/derby/VMCLIDB already exists.
    (...)
    DAEMON 15:03:27,353 (daemon) FMM16020W A valid Derby database connection was not found.

    When trying to create a new VMCLIDB by starting the vmcli daemon it fails with similar messages.

    Work-around: To recreate file service.properties, use one of these options:
    • Copy file service.properties from the latest VMCLI database backup located on the file system.
      The default location is install_dir/derby_backups.
      For example on Linux this is /opt/tivoli/tsm/tdpvmware/common/derby_backups.
      Copy back the properties file from the backup to the broken VMCLI database.
    • Use file service.properties from a newly created VMCLI database:
      1. Rename the broken VMCLI database directory (default: /home/tdpvmware/tdpvmware/VMCLIDB)
      2. Then restart the vmcli daemon to create a new VMCLI database:
        /etc/init.d/vmcli restart
      3. Stop the daemon:
        /etc/init.d/vmcli stop
      4. Copy the service.properties file from the newly created VMCLI database to the broken database directory.
      5. Delete the new VMCLI database.
      6. Rename the broken VMCLI database back to the original name.
      7. Start the vmcli daemon:
        /etc/init.d/vmcli start
    (86868)

  • Warning message FMM2094W does not list all possible causes of the situation

    The following warning message is issued when an operation interferes with an IBM Spectrum Protect™ Snapshot for VMware backup process:

    FMM2094W A file layout change was detected during the backup of the virtual machine 'vm_name', this change was probably caused by a vMotion operation.
    The virtual machine is excluded from the backup and cannot be restored from this backup.

    However, there are other operations that may have the same effect if they coincide with the backup process.
    Examples for such operations are:
    • Virtual machine snapshots
    • Virtual machine power state changes (Power off, Power on, Suspend, Resume, Restart)
    See the IBM Tivoli Storage FlashCopy Manager for VMware v4.1 Installation and User's Guide, section "Migrating virtual machines and Tivoli Storage FlashCopy Manager for VMware hardware snapshot interoperability" for more details.

  • Duplicate informational messages

    During IBM Spectrum Protect™ Snapshot for VMware operations, a customer may see some duplicate informational messages in the log files.
    Example for a message that can occur multiple times:
    FMM6555I Selected snapshot backup with ID '<BACKUP_ID>'.

    There is no work-around. The duplicate messages can be ignored. (82060)

  • A new backup fails if another backup is currently in use by other IBM Spectrum Protect™ Snapshot for VMware operations on DS8000 storage.
    Preconditions of this problem:
    1. The virtual machines to protect reside on DS8000 storage.
    2. At least two sets of target volumes are used.
    3. Virtual disks of an available backup are attached to another virtual machine or a restore / TSM backup of an available backup is currently running.

    A new backup will fail with message:
    FMM1039E Aborting backup. Another backup with ID '<backup_id>' is already mounted on backup server: '<backup_server>'.

    Work-around: In case virtual machines of another backup are attached to another virtual machine, detach these backups. Then restart the new backup.
    In case a restore / TSM backup of an available backup is currently running, wait for it to finish. Then restart the new backup.

  • If a datastore is shared between different datacenters, IBM Spectrum Protect™ Snapshot for VMware will randomly pick one of the datacenters and only backup the VMs which are located on the datastore and registered in this datacenter of the ESX host of the datastore. Due to this behaviour the VMs from same datastore that are registered in the other datacenter will not be restorable.

    Work-around: The customer has to avoid vSphere environment configurations where a datastore is used in different datacenters, if he plans to backup this datastore with FCM for VMware.
    (33860)

  • When restoring a virtual machine not to its original datastore but to a different datastore a misleading error message can be generated. The following message is generated if a virtual machine with the same name already exists in the datacenter:
    "ERROR FMM0416E The '<virtual_machine_name>' virtual machine cannot be restored to the '<datastore_name>' datastore using the '<virtual_machine_name>' VM name. A VM with the same name already exists in the datastore."
    The correct error message is:
    "ERROR FMM0416E The '<virtual_machine_name>' virtual machine cannot be restored to the '<datastore_name>' datastore using the '<virtual_machine_name>' VM name. A VM with the same name already exists in the vCenter."
    (43642)

  • When trying to run an instant restore, while the background copy on the storage system is still in progress for this backup id, the instant restore fails with the error message:

    "ERROR FMM0552E A backup with the '<BID>' ID cannot be found."

    Work-around: Wait until the background copy process has completed by monitoring the USABILITY_STATES in the inquire_detail command output:

    ./vmcli -f inquire_detail -t FCM

    The output will show the BACKGROUND_MONITOR_PENDING state in his USABILITY_STATES until the background copying process ends. Once the state is reset, rerun the instant restore operation. (87584)

  • IBM Spectrum Protect™ Snapshot for VMware is configured to manage a subset of datacenters, by not selecting the "Manage all" check box in the domain configuration wizard. In this scenario, if one of the managed datacenters is deleted in the vCenter server, all successive backups fail with the following error message:

  • ERROR FMM0365E The datacenter with the name '<name of deleted datacenter>' was not found.

    The error occurs even if the backup does not involve any virtual machine or datastore from the deleted datacenter. (89779)

    Work-around: Remove the deleted datacenter from the list of managed datacenters in the IBM Spectrum Protect™ Snapshot for VMware domain configuration.

  • When creating an IBM Spectrum Protect™ Snapshot for VMware backup schedule with offload to IBM Spectrum Protect™, there is a problem when selecting a large amount of virtual machines. The critical amount of selected VMs is depending on the name length of the VMs and the containing data stores. IBM Spectrum Protect™ Snapshot for VMware is creating content to be stored in the VMCLI Derby database that can became too large for the specific data base column. The result is a failure in the backup, attach or detach operation with the following exeception can be found in the vmcli traces: "com.ibm.storage.vmcli.exceptions.VmcliDBException: java.sql.SQLException: An IOException was thrown when reading a 'java.lang.String' from an InputStream" with the cause "org.apache.derby.client.am.SqlException: Java exception: 'A truncation error was encountered trying to shrink LONG VARCHAR '<stream-value>' to length 32700.: org.apache.derby.iapi.services.io.DerbyIOException'".

    Workaround: Create multiple backup schedules with less virtual machines selected. (119039)

  • When using the GUI ConfigWizard for the first time after the installation to create a Device Class with type "IBM System Storage DS8000" there is a problem with the check for the Target Set Definition file: If a wrong value is entered the expected warning message FMM8463W does not appear and the entry field for the file is not marked red. The behavior varies dependent on what kind of (wrong) file name is entered:
    1. if any text but without a / at the beginning is entered, no warning appears and the "Acknowledge Warnings and Save" button is not activated. User cannot proceed until a valid file name is entered or any text with a / at the beginning (which leads to variant 2 following)
    2. if any text with a / at the beginning is entered, no warning appears but  the "Acknowledge Warnings and Save" button IS activated. User can proceed by clicking this button which results in an invalid file name was accepted by the configuration wizard.

    However, if you revisit the Device class configuration again (edit mode) everything works as expected:
    1. if any text without a / at the beginning is entered, an error FMM8332E appears and the "Acknowledge Warnings and Save" button is not activated
    2. if any text with a / at the beginning is entered, no warning appears but  the "Acknowledge Warnings and Save" button IS activated.

  • Limitations releated to XIV Multi-Site Mirroring environments:
    • FlashCopy Manager fails to restore snapshots if the volumes containing the application data are mirrored into two or more destinations with XIV Multi-Site Mirroring.
    • FlashCopy Manager fails to create snapshots of mirrored volumes if the application data is mirrored into two or more destinations with XIV Multi-Site Mirroring. Snapshots of local source volumes can be created.
    In both scenarios FlashCopy Manager fails with an error message containing this text: "VOLUME_HAS_MULTIPLE_MIRRORS volume has multiple mirrors, operation not allowed or target must be specified"

    Work-around: Remove all mirror relations manually, before triggering a restore operation with FCM.
    To create a snapshot of a remote mirror, ensure that only one mirror of the volumes containing the application data exists.

  • For IBM Spectrum Protect™ Snapshot to work in combination with dynamic target allocation (SVCDTA) device classes, the name of the SAN Volume Controller production source volumes must not exceed the length of 27 characters.

    An IBM Spectrum Protect Snapshot backup involving a volume with a longer name fails with the following SVC error message:
    CMMVC5738E The argument contains too many characters.

    There is no work-around to allow source volume names of length larger than 27 characters.
    The sample below shows however that a volume name smaller than 27 characters can be practicable.
     
    Example:
    Given the sample input source volume name „esxcet1_vol1“, IBM Spectrum Protect Snapshot SVC DTA adapter would generate the following object names in SAN Volume Controller:

    Name of the dynamic allocated target volume for the snapshot backup operation:
    esxcet1_vol1_A0ITWN5DGH_ITWN6A1F_1
    Name of the duplicate target volume used for the mount operation:
    esxcet1_vol1_A0ITWN5DGH_ITWN6A1F_1_dup
    Name of the FlashCopy mapping:
    TSM_A0IO1BN9X6_A0IW39B9GG_IW39J8A6_esxcet1_vol1
    Name of the cosistency group
    TSM_A0IO1BN9X6_A0IW39B9GG_IW39J8A6  



Known Issues and Limitations Specific to Off-loaded Backup to a IBM Spectrum Protect™ Server

  • A backup may stay attached after disrupting an off-loaded backup to an IBM Spectrum Protect™ server

    If an off-loaded backup to an IBM Spectrum Protect™ server was disrupted, for example by an unexpected system shutdown, the attached backup is not cleaned up. This will not interfere with any additional attach and restore operations, but it might cause a problem when the target set of this backup is to be reused for a new backup. In this case the target set cannot be reused when the backup is still attached.

    Work-around: To check if a backup is still in attached state issue the following command:
    ./vmcli -f inquire_detail -t FCM

    In the command output, look for the values of "#PARAM USABILITY_STATES".
    If you know that there is no backup to an IBM Spectrum Protect™ server currently running, but find a backup with the USABILITY_STATE TAPE_BACKUP_IN_PROGRESS and MOUNTED, and if no VMs of this backup are currently attached you may clean up the attached state by running this command:
    vmcli -f detach -B backupid

    This command detaches all VMs and datastores that are currently attached to an ESX host and resets the USABILITY_STATE for the corresponding backup. If the backup was not completely attached because of a failed attach operation, this will do a cleanup of the partially attached or unattached backup and reset its USABILITY_STATE.
    (86571)

  • For performing off-loaded backups to a IBM Spectrum Protect™ Snapshot server, IBM Spectrum Protect™ for Virtual Environments is employed. Known problems and limitations of this product may also apply to off-loaded backups in combination with IBM Spectrum Protect™ Snapshot for VMware. They are documented at http://www.ibm.com/support/docview.wss?uid=swg21640744.

Resolved Issues with FlashCopy Manager 4.1.1.0
  • Datastore restore wizard provides registration checkbox for ineligible VMs (99029)
Issues and Limitations specific to FlashCopy Manager 4.1.1.0 and later modifications
  • When upgrading from a previous FlashCopy Manager version, when the GUI based configuration wizard opens for the first time, the password fields show empty even if the password file actually includes the passwords (from a previous installation).

    Work-around: Re-enter the password on the vCenter credentials page, then click through all remaining pages with the Next button and click Finish on the summary page so that the profile is saved. When revisiting the configuration wizard all passwords are present again. (76124)
  • Launching the VMware vSphere Client plug-in version of the FlashCopy Manager GUI right after installation creates a wrong symbolic link to the actual profile file. As a consequence, the configuration wizard refuses to open and the product is not operational.

    Work-around: To avoid this issue, either use the console based configuration (setup.sh) or the stand-alone version of the Web GUI (which opens automatically after installation) for the initial configuration but not the plug-in version. You can run the plug-in version after that at anytime without an issue.

    If you already run into the issue, remove the wrong link (rm -rf /opt/tivoli/tsm/tdpvmware/fcm/profile) manually, then run the console or Web GUI based configuration wizard (NOT the plug-in version!) (75343)
  • In a remote mirrored environment, the attempt to restore or attach virtual machines on the remote site (secondary site) fails with the vSphere API error 'fault.VmfsAlreadyMounted'.

    Scenario leading to this issue: In case of a fail over the remote site replica of a mirrored datastore is manually mounted to a secondary site ESX host in which the datastore is not resignatured.

    The attempt to restore virtual machines from a Flashcopy Manager hardware snapshot backup of this datastore fails with the said error message because of the mounted (non-resignatured) datastore is considered to be a snapshot copy itself since its meta data still refers to the primary site LUNs. According to the VMware KB entry 2082643 mounting / resignaturing a second snapshot copy to the same host is not supported in principle.

    The issue occurs for instant restore, normal restore, and disk attach operations.

    Work-around: In the context of a fail over to the remote (secondary) site, when mounting datastores manually (in absence of SRM), ensure that the mounted datastores are resignatured in that moment. Note: For Site Recovery Manager (SRM) protected datastores this is done automatically so you will not run into this issue when using SRM. (62234)
  • The warning message "Internet Explorer 12+ is not supported" might show up when launching the VMware vSphere Client on a machine with Internet Explorer 11.* installed as the native web browser.

    Work-around: Close the warning message - the product works as expected anyway. (75562)
  • When using the vmcli commands managed_capacity or used_capacity (deprecated) in combination with the -t FCM option, the command will fail.

    Work-around: Run the command without the option -t FCM - both commands are implicitly related to FCM anyway. (75463)
  • The warning "FMM8682W The current user ID does not have permissions to create or update tasks in the vCenter" might occur during backup or restore operations.

    The backup or restore operation is still successful but ends with rc=1 (warning). The warning message relates to a failed task progress update in the task progress view so is not related to the actual operation it represents. As a consequence, the task status in the task progress view remains in status "in progress" for said operation. The warning is not induced by lack of permissions as the message indicates.

    Work-around: If the warning occurs sporadically, it can be ignored. If the warning occurs on every operation, please check for your user permissions: Is it allowed to create and update tasks? (75185)
  • In the Tivoli Storage FlashCopy Manager for VMware and Tivoli Storage Manager Data Protection for VMware GUIs, the configuration area has a panel that shows a list of managed datacenters currently being managed in the GUI domain. When this list is modified, for example by removing some data centers out of the list and after that saving the new configuration, the removed data centers do still show up in the source panel of the backup wizard when a new backup schedule is created.

    Work-around: After changing the managed data center list, log out and then log back into the GUI so that the managed data center list is refreshed. (74582)
  • In the Tivoli Storage FlashCopy Manager for VMware and Tivoli Storage Manager Data Protection for VMware GUIs, the configuration area has a panel that shows a list of managed datacenters currently being managed in the GUI domain. On this panel there is a checkbox named "Manage all available datacenters" that when checked reflects that the GUI will allow all available datacenters to be managed, including any new datacenters that are subsequently created. When this checkbox is not selected, then only the specific list of datacenters is managed by the GUI. Normally this specific list is a subset of all the available datacenters.

    The problem occurs in the unusual case where the domain configuration begins with the "Manage all available datacenters" checkbox unchecked, but the managed datacenter list is set to be the set of all datacenters currently available in the vCenter. Later, when this panel is opened again, the GUI detects that all available datacenters are being managed, and incorrectly shows the "Manage all available datacenters" checkbox as being checked. This is incorrect because in fact there is a specific list of managed datacenters set, and any new datacenters will not be automatically managed. This has no impact at the time the panel is viewed, but in the future if a new datacenter is added to the vCenter that datacenter will not automatically be managed by the GUI.

    Work-around: If seeing an issue where a datacenter is not managed and yet the Manage all datacenters checkbox is selected, then reset the checkbox state using the following procedure:

    1) Open the Managed Datacenter (GUI Domain) panel, temporarily unselect the checkbox, and move one datacenter to the available list and save.
    2) Re-open the the panel, select the Manage all available datacenters checkbox, and save.

    The result is that the checkbox is actually set, so all currently available datacenters and any future added datacenters will be managed in the GUI. (75070)
  • In the case of manually editing the FCM profile (which is not recommended) and accidentally removing the COPYSERVICES_HARDWARE_TYPE in a device class section, the GUI configuration wizard only shows the previous device class sections of the profile. When the configuration is continued and finished so that the profile is saved, the invalid device class and all following device classes are removed from the profile.

    Work-around: When a new profile is saved, a copy of the original profile is created in the config folder where the profile is located. The profile is normally stored in <tdpvmware user home>/tdpvmware/config and has the name profile.bck. Copy back this previous version of the profile and correct the invalid DEVICE_CLASS section. As an alternative, after copying back the profile, you can run the console based configuration wizard which handles the invalid DEVICE_CLASS gracefully and retrofits the missing parameter. (73319)
  • The help output of the InstallAnywhere installer (<installername>.bin --help) includes an option

    -r <path_to_generate_response_file> that generates a response file.

    The response file generated from the option "-r" cannot be used for silent installation.

    Work-around:
    Use the installer property option ("-DRECORDFILE") to generate an appropriate response file for silent installation. This installer property is explained in the documentation. (73087)
  • For remote site restores in combination with XIV storage, Tivoli Storage FlashCopy Manager issues a misleading error message when either of the following situations occur (88405):

    - An invalid server address is specified in the FlashCopy Manager profile for the remote site XIV Storage System.
    - The remote site XIV Storage System is temporarily not reachable on the network:

    #ERROR FMM18103E The XIV device adapter was not able to collect the information for the DS_XC1_fcmcli1_XIV_1 consistency group. The following error message was generated by the IBM XIV Storage System: {1}.

    Work-around: Ensure that the server address for the XIV remote system is correctly configured in the Tivoli Storage FlashCopy Manager profile and that it is reachable on the network. When the profile is configured with the console or GUI-based configuration wizard, this configuration is automatically verified when the password for the storage device is entered.
  • When a new backup schedule is created the information of the selected device class gets overwritten with the default device class when stepping back to the source VM selection page of the configuration wizard. The problem only occurs when the user goes back from the summary page to the beginning of the backup wizard to revisit the device class selection page, which is by then reinitialized with the default device class, even though some information about the previously selected device class still shows up on the page. (109633)

    Work-around: Select the desired device class again, if it has been overwritten with the default device class when revisiting the device class selection page.
  • When the product is installed without a license file (FTP image), the GUI configuration wizard of IBM Spectrum Protect™ Snapshot for VMware issues a misleading error message when entering the vCenter credentials: (110098)

    "FMM2012E The specified user name or password is not correct. authentication failed"

    So it appears like the entered credentials are wrong, however, in fact the error message is triggered by the lack of the license file.

    Work-around: Ensure that the license file is deployed.
  • For SVC / Storwize remote mirrored environments using Global Mirroring with Change Volumes, an instant restore on the secondary site might fail with the following error message:

    FMM5667E A Flashcopy Restore operation is in progress between source volume X and target volume Y. Please retry this command after the restore operation completes.

    Workround:
     Once you ensured that there is in fact no other restore operation in progress, remove the Change Volumes from the Global Mirror relationship or remove the Global Mirroring relationship itself, then retry the operation.
Resolved Issues with FlashCopy Manager 4.1.2.0
  • When upgrading from a previous FlashCopy Manager version, when the GUI based configuration wizard opens for the first time, the password fields show empty even if the password file actually includes the passwords (from a previous installation). (76124)
  • Launching the VMware vSphere Client plug-in version of the FlashCopy Manager GUI right after installation creates a wrong symbolic link to the actual profile file. As a consequence, the configuration wizard refuses to open and the product is not operational. (75343)
  • The warning message "Internet Explorer 12+ is not supported" might show up when launching the VMware vSphere Client on a machine with Internet Explorer 11.* installed as the native web browser. (75562)
  • When using the vmcli commands managed_capacity or used_capacity (deprecated) in combination with the -t FCM option, the command will fail. (75463)
  • The warning "FMM8682W The current user ID does not have permissions to create or update tasks in the vCenter" might occur during backup or restore operations. (75185)
  • In the Tivoli Storage FlashCopy Manager for VMware and Tivoli Storage Manager Data Protection for VMware GUIs, the configuration area has a panel that shows a list of managed datacenters currently being managed in the GUI domain. When this list is modified, for example by removing some data centers out of the list and after that saving the new configuration, the removed data centers do still show up in the source panel of the backup wizard when a new backup schedule is created. (74582)
  • In the case of manually editing the FCM profile (which is not recommended) and accidentally removing the COPYSERVICES_HARDWARE_TYPE in a device class section, the GUI configuration wizard only shows the previous device class sections of the profile. When the configuration is continued and finished so that the profile is saved, the invalid device class and all following device classes are removed from the profile. (73319)
  • In the Tivoli Storage FlashCopy Manager for VMware and Tivoli Storage Manager Data Protection for VMware GUIs, the configuration area has a panel that shows a list of managed datacenters currently being managed in the GUI domain. On this panel there is a checkbox named "Manage all available datacenters" that when checked reflects that the GUI will allow all available datacenters to be managed, including any new datacenters that are subsequently created. When this checkbox is not selected, then only the specific list of datacenters is managed by the GUI. Normally this specific list is a subset of all the available datacenters. (75070)

Issues and Limitations specific to FlashCopy Manager 4.1.2.0 and later modifications
  • An SRM version compatible with vSphere 6 was not available through the VMware beta program so coexistence with SRM on vSphere 6 is not verified.
  • For vSphere 6 environments, the Data Protection for VMware vCenter plug-in is not supported.

    Work-around:
    Use the stand-alone web browser GUI option or the Extension in the Web Client instead.
  • If you use FCM to back up a virtual machine with storage policies assigned to itself or any of its virtual disks, these policies are not preserved after a restore operation. After executing a restore operation the VM is assigned to the default VM storage policy. (86142)

    Work-around: Reassign the proper storage policies after the restore is completed.
  • When you select a language other than English for the product installation, in the vSphere Protection GUI information page, a default method to access the GUI is not selected. Using the English local, the option "Enable access to the GUI by web browser" is preselected. You cannot proceed until you select at least one option. (90086)

    Work-around: Select at least one option to access the GUI.
  • When you install Tivoli Storage Manager for Virtual Environments V7.1.2 to a system where Tivoli Storage FlashCopy Manager for VMware V4.1.1 or earlier is installed, the outdated version of the installation is automatically deactivated.

    You can reactivate the installation by upgrading to Tivoli Storage FlashCopy Manager V4.1.2. However, the former installation of Tivoli Storage Manager for Virtual Environments V7.1.2 resets the VCENTER_SERVER and VCENTER_SERVER_USER parameters to an empty value in the Tivoli Storage FlashCopy Manager profile. Both parameters are mandatory to run Tivoli Storage FlashCopy Manager.

    Work-around: After the upgrade of Tivoli Storage FlashCopy Manager to V4.1.2, run the configuration wizard again to add the missing parameters. You must use the console-based and not the GUI-based configuration wizard as the GUI-based configuration wizard requires the missing parameters.
  • The attempt to create a combined FlashCopy Manager / TSM for Virtual Environments backup (i.e. FCM backup with offload to TSM) fails for virtual machines with RDM disks attached on vSphere 6.0 if the backup mode is SNAPSHOT_EXCL_MEM or SNAPSHOT_INCL_MEM. The FCM backup succeeds but the offload backup to TSM fails with (103191):

    ERROR FMM0129E The following error was returned by the vSphere(TM) API: 'Invalid configuration for device '0'.'.

    Work-around: a) Use backup modes SUSPEND or ASIS or b) perform direct backups to TSM or c) exclude VM with RDM disks from the backup.

Resolved Issues with IBM Spectrum Protect™ Snapshot for VMware 4.1.3.0
  • For remote site restores in combination with XIV storage, Tivoli Storage FlashCopy Manager issues a misleading error message when either of the following situations occur (88405):

    - An invalid server address is specified in the FlashCopy Manager profile for the remote site XIV Storage System.
    - The remote site XIV Storage System is temporarily not reachable on the network:

    #ERROR FMM18103E The XIV device adapter was not able to collect the information for the DS_XC1_fcmcli1_XIV_1 consistency group. The following error message was generated by the IBM XIV Storage System: {1}.
Issues and Limitations specific to IBM Spectrum Protect™ Snapshot for VMware 4.1.3.0 and later modifications
  • When creating a new device class with type "IBM System Storage DS8000" using the GUI Configuration Wizard, the validation for the Target Set Definition file is erroneous: If a wrong value is entered the expected warning message FMM8463W does not appear and the corresponding entry field is not marked red. The behavior varies dependent on what kind of (wrong) file name is entered: (103521)

    1. If any text is entered but without a "/" at the beginning, no warning appears and the "Acknowledge Warnings and Save" button is not activated. You cannot proceed until a valid file name is entered or any text with a "/" at the beginning (leading to variant 2 following)
    2. If any text with a "/" at the beginning is entered, no warning appears but the "Acknowledge Warnings and Save" button is in fact activated. You can proceed by clicking this button but this results in an invalid file name is written to the profile.

    When revisiting the Configuration wizard and editing the device class again, the field is validated correctly. You will receive an error or warning message if appropriate.

    Work-around: Carefully enter a valid file name when creating the new device class or revisit the configuration page right after in order to get the entered value correctly validated.
  • Limitation for environments on IBM System Storage SAN Volume Controller (SVC), IBM Storwize® family, IBM Flex System™ v7000, or IBM FlashSystem™ V840 where source volumes of a FlashCopy are mirrored by using the ability of the SVC to create volumes with two copies, and where the copies are in two different SVC storage pools:
    When using dynamic allocation of FlashCopy target volumes in such environments the SVC storage pool for the target volumes of the FlashCopy is not automatically determined.

    Work-around: Specify the SVC storage pool for the target volumes explicitly via parameter SVC_POOLNAME in the DEVICE_CLASS section of the IBM Spectrum Protect Snapshot profile.
Resolved Issues with IBM Spectrum Protect™ Snapshot for VMware 4.1.4.0
  • The attempt to create a combined FlashCopy Manager / TSM for Virtual Environments backup (i.e. FCM backup with offload to TSM) fails for virtual machines with RDM disks attached on vSphere 6.0 if the backup mode is SNAPSHOT_EXCL_MEM or SNAPSHOT_INCL_MEM. The FCM backup succeeds but the offload backup to TSM fails with (103191):

    ERROR FMM0129E The following error was returned by the vSphere(TM) API: 'Invalid configuration for device '0'.'.

Issues and Limitations specific to IBM Spectrum Protect™ Snapshot for VMware 4.1.4.0 and later modifications
  • In IBM Spectrum Protect™ Snapshot only installations (not combined with IBM Spectrum Protect™ for Virtual Environments), choosing the "Show back up history" action or the "Show all history" action from the context menu of a selected entry in a Backup Status report results in a dialog that appears in the loading state forever (109796).

    Work-around: These actions should not be available in IBM Spectrum Protect™ Snapshot only installations.
  • In the IBM Spectrum Protect™ Snapshot for Vmware GUI configuration wizard, for SVCDTA DEVICE_CLASSes there is an input field labeled SSH key file for entering the path to the SSH key file that is used by the storage adapter to authenticate against the SVC storage device. The default of this parameter is ~/.ssh/svc_sshkey and should fit in most cases. The ~ symbol refers to the home directory of the instance user in which context the product is executed later. However, this might be mixed up with the SVC user (for that the referred key was imported) that is prompted just above the SSH keyfile parameter (label User Name). So the presentation of these parameters on the GUI panel has the potential to be misinterpreted. (109669)

    Work-around: Ensure that the user name entered under the label User Name is in fact the SVC user for that the key file specified under the label SSH key file has been imported on the SVC storage device.

[{"Product":{"code":"SS36V9","label":"Tivoli Storage FlashCopy Manager"},"Business Unit":{"code":"BU058","label":"IBM Infrastructure w\/TPS"},"Component":"FlashCopy Manager for VMware","Platform":[{"code":"PF016","label":"Linux"}],"Version":"4.1","Edition":"","Line of Business":{"code":"LOB26","label":"Storage"}}]

Document Information

Modified date:
17 June 2018

UID

swg21651408