Delta Bundle Upgrade#

Important

  • When upgrading from before release 25.2, it is recommended that commands in installation steps are run in a terminal opened with the screen command. Note that after this upgrade, the screen command is replaced with the tmux command.

  • When upgrading from 25.2, it is recommended that commands in installation steps are run in a terminal opened with the tmux command.

Download patch script and checks#

Unified-Node-Cluster Modular-Node-Cluster Single-Unified-Node

Note

It is recommended that the file download is done prior to the maintenance window.

Download the following file to the media folder on the node relevant to your topology:

25.3-Delta-Bundle-patch.script

Topology

The node on which to download the file

Unified-Node-Cluster

Download the file on the primary unified node

Modular-Node-Cluster

Download the file on the primary application node

Single-Unified-Node

Download the file on the single node

Note

Unified-Node-Cluster Modular-Node-Cluster

This note is relevant only to a unified node cluster topology and a modular node cluster topology since, in a single unified node environment, the application and database primary is on the same node.

To determine where the current primary application and database node is running:

  1. Log in on any node.

  2. Run the following command: $ cluster run application cluster primary role application

  3. Record the node entry where is_primary: true, for example:

---------- VOSS-UN-1, ip=192.168.100.3, role=webproxy,application, loc=cpt

is_primary: true

Verify SHA256 checksum#

Unified-Node-Cluster Modular-Node-Cluster Single-Unified-Node

To verify SHA256 checksum for the patch, run the following command on the node the script was downloaded to:

system checksum media/25.3-Delta-Bundle-patch.script

The expected output is the value in 25.3-Delta-Bundle-patch.script.sha256, available in the download location.

Pre-installation version check#

Unified-Node-Cluster Modular-Node-Cluster Single-Unified-Node

To check the version pre-install:

  1. Log in to the Admin Portal GUI.

  2. Verify the information contained in About > Version > Release.

    The release version must be 25.1 or 25.2

Pre-installation and health checks#

Unified-Node-Cluster Modular-Node-Cluster Single-Unified-Node

  1. Place the system in maintenance mode and suspend any scheduled transactions.

    On the primary node of the system, run: cluster maintenance-mode start

    Scheduled transactions that are in progress will be allowed to complete.

    For details on this command, refer to the System Maintenance Mode topic in the Platform Guide.

    To verify the maintenance mode status, run: cluster maintenance-mode status

  2. Verify network connectivity, disk status, NTP.

    1. Run the following commands:

      cluster check

      cluster status

    2. Check status - details for this step depend on your topology:

      Unified-Node-Cluster Modular-Node-Cluster

      On the primary application node, run:

      cluster run all diag disk

      Single-Unified-Node

      On the single node, run: diag disk

    3. Clean up disk space, if required.

      If there is any sign that the paths below are over 80% full, a clean-up is needed on every node to avoid risk of, for example, full logs occurring during upgrade. Clean-up steps are indicated next to the paths:

      Path

      Resolution

      /

      Contact VOSS Support if over 80%

      /var/log

      Run cluster run all log purge to execute on every node

      /opt/platform

      Log in to each node individually and remove any unnecessary files from the /media directory

      /tmp

      Reboot

  3. Shut down servers and take snapshots from VMWare or Azure VHD, as applicable - details for this step depend on your topology:

    Unified-Node-Cluster

    On the primary unified node, run: cluster run notme system shutdown

    After one minute, run: system shutdown

    Modular-Node-Cluster

    On the primary application node, run: cluster run notme system shutdown

    After one minute, run: system shutdown

    Single-Unified-Node

    On the single node, run: system shutdown

Delta or Patch Bundle installation#

Unified-Node-Cluster Modular-Node-Cluster Single-Unified-Node

  1. Install the patch. The details for this step depends on your topology:

    Unified-Node-Cluster

    On the primary unified node, run:

    screen

    app install media/25.3-Delta-Bundle-patch.script

    Modular-Node-Cluster

    On the primary application node, run:

    screen

    app install media/25.3-Delta-Bundle-patch.script

    Single-Unified-Node

    On the single node, run:

    screen

    app install media/25.3-Delta-Bundle-patch.script

    Note

    Before the patch installation starts, the following prompts display - in each case, use the actions indicated for each prompt to remove the prompt:

    Prompt:

    Continue with the installation.

    Action: Append the --force parameter to remove this prompt.

    Prompt:

    Delete or keep the patch script in the media directory after installation.

    Action: Append the delete-on-success parameter with a yes|no value to the command to remove this prompt.

    To remove all prompts, use the command and parameters:

    app install media/25.3-Delta-Bundle-patch.script delete-on-success yes --force

  2. Close screen.

Post-installation, security and health checks#

Unified-Node-Cluster Modular-Node-Cluster Single-Unified-Node

  1. Verify that the cluster is stable after the upgrade. Run the following on the primary node:

    cluster check

    Any errors that are reported need to be corrected prior to resuming the upgrade.

  2. The message: *** Reboot Required - New Kernel Installed vmlinuz-x.xx.x-xxx-generic *** may be displayed at the bottom after the upgrade.

    Reboot the cluster. The details for this step depend on your topology:

    Unified-Node-Cluster Modular-Node-Cluster

    To reboot all nodes, run: cluster run notme system reboot

    If the following node messages display, these can be ignored: <node name> failed with timeout

    Then run: system reboot

    Single-Unified-Node

    On the single node, run: system reboot

    Note

    Since all services will be stopped, this step takes some time.

  3. End the system maintenance mode. On an application node of the system, run:

    cluster maintenance-mode stop

    Verify maintenance mode status, run: cluster maintenance-mode status

    For details on this command, refer to the System Maintenance Mode topic in the Platform Guide.

  4. Run cluster check.

    Inspect the output for warnings and errors. You can also use cluster check verbose to see more details, for example, to check that avx is enabled.

    For troubleshooting and resolutions, also refer to the Health Checks for Cluster Installations Guide and the Platform Guide. Contact VOSS Support for assistance, if required.

  5. Verify the upgrade:

    1. Log in on the Admin Portal, and check version details in About > Version.

      If your web browser can’t open the user interface, clear your browser cache before trying to open the interface again.

    2. Confirm that versions are upgraded (where XXX is the release version).

      • Release should display XXX

      • Platform version should display XXX

    3. Check that themes on all roles are set correctly.