InfoScale™ 9.0 Storage Foundation Cluster File System High Availability Configuration and Upgrade Guide - Solaris

Complete the following tasks before you perform the upgrade:

• Review the Veritas InfoScale 9.0 Release Notes for any late-breaking information on upgrading your system.

• Review the Veritas Technical Support website for additional information:

  https://www.veritas.com/support/en_US.html

• You can configure the Veritas Telemetry Collector while upgrading, if you have do not already have it configured. For more information, refer to the About telemetry data collection in InfoScale section in the Veritas Installation guide.

• Perform the following system-level settings:

  • Set diag-level to min to perform the minimum number of diagnostics when the system boots. Depending on the configuration of your systems you may want to turn it on after you perform the upgrade.

    {1} ok setenv diag-level min

     diag-level=min

  • Set auto-boot? to false. For tight control when systems restart, set this variable to false. Re-enable this variable after the upgrade.

    {1} ok setenv auto-boot? false

    auto-boot?=false

  • Deactivate cron to make sure that extraneous jobs are not performed while you upgrade the systems.

    Solaris 11:

    # ps -ef | grep cron
    # kill cron pid
    # svcadm disable svc:/system/cron:default

• If zones are present, make sure that all non-global zones are booted and are in the running state before you use the Veritas InfoScale product installer to upgrade the Storage Foundation products in the global zone so that any packages present inside non-global zones also gets updated automatically.

  If the non-global zone has previous version of VCS packages (VRTSperl, VRTSvlic, VRTSvcs, VRTSvcsag, VRTSvcsea) already installed, then during upgrade of the VCS packages in global zone, packages inside non-global zone are automatically upgraded if the zone is in running state. If non-global zones are not in running state, you must set the Veritas publisher inside the global zone. You also must attach the zone with - u option to upgrade the SFHA packages inside non-global zone. If previous version of VRTSvxfs, and VRTSodm packages are installed inside non- global zone, they must be uninstalled manually prior to the upgrade. Once the packages in global zone are upgraded, VRTSvxfs and VRTSodm must be installed manaully inside non-global zone.

• Make sure that the administrator who performs the upgrade has root access and a good knowledge of the operating system's administration.

• Make sure that all users are logged off and that all major user applications are properly shut down.

• Make sure that you have created a valid backup.

• Ensure that you have enough file system space to upgrade. Identify where you want to copy the packages, for example /packages/Veritas when the root file system has enough space or /var/tmp/packages if the /var file system has enough space.

  Do not put the files under /tmp, which is erased during a system restart.

  Do not put the files on a file system that is inaccessible before running the upgrade script.

  You can use a Veritas-supplied disc for the upgrade as long as modifications to the upgrade script are not required.

  If /usr/local was originally created as a slice, modifications are required.

• Unmount all the file systems not on the root disk. Comment out their entries in /etc/vfstab. Stop the associated volumes and deport the associated disk groups. Any file systems that the Solaris operating system or Storage Foundation assumes should be in rootdg but are not, must be unmounted, and the associated entry in /etc/vfstab commented out.

• For any startup scripts in /usr/sbin/svcadm disable, comment out any application commands or processes that are known to hang if their file systems are not present.

• Make sure that the current operating system supports version 9.0 of the product. If the operating system does not support it, plan for a staged upgrade.

• Schedule sufficient outage time and downtime for the upgrade and any applications that use the Veritas InfoScale products. Depending on the configuration, the outage can take several hours.

• Any swap partitions not in rootdg must be commented out of /etc/vfstab. If possible, swap partitions other than those on the root disk should be commented out of /etc/vfstab and not mounted during the upgrade. The active swap partitions that are not in rootdg cause upgrade_start to fail.

• Make sure that the file systems are clean before upgrading.

• Arctera recommends that you upgrade VxFS disk layouts to a supported version before installing VxFS 9.0. Unsupported disk layout versions can be mounted for the purpose of online upgrading in VxFS 9.0. You can upgrade unsupported layout versions online before installing VxFS 9.0.

  See Upgrading disk layout versions.

• Upgrade arrays (if required).

  See Upgrading the array support.

• To reliably save information on a mirrored disk, shut down the system and physically remove the mirrored disk. Removing the disk in this manner offers a failback point.

• Determine if the root disk is encapsulated.

• Make sure that DMP support for native stack is disabled (dmp_native_support=off). If DMP support for native stack is enabled (dmp_native_support=on), the installer may detect it and ask you to restart the system.

• If you want to upgrade the application clusters that use CP server based fencing to version 7.3.1 and later, make sure that you first upgrade VCS or SFHA on the CP server systems to version 7.3.1 and later. And then, from 7.3.1 onwards, CP server supports only HTTPS based communication with its clients and IPM-based communication is no longer supported. CP server needs to be reconfigured if you upgrade the CP server with IPM-based CP server configured.

  For instructions to upgrade VCS or SFHA on the CP server systems, refer to the relevant Configuration and Upgrade Guides.