HP StoreAll OS 6.5.1 Upgrade Guide

Table 1 Prerequisites checklist for the StoreAll upgrade (continued)

Step

completed?DescriptionStep

IMPORTANT: All the StoreAll nodes must be at the same release.

• If you are running a version of StoreAll earlier than 6.1.4, perform the upgrade as

described in “Cascading upgrades” (page 18).

• If you are running StoreAll 6.1.4 or later, proceed to step 2.

Verify that the / partition contains at least 1 GB and the /local partition contains at least

4 GB for the upgrade by entering the following commands:

pdsh -a [-f MaxStartupValue] df -h /

pdsh -a [-f MaxStartupValue] df -h /local

Ensure no tasks are running. Run ibrix_task -l to list active tasks. To view completed

tasks, enter ibrix_task –l –c. Stop any active remote replication, data tiering, or

rebalancer tasks running on the cluster, so that the tasks will not be terminated by the

upgrade script. Use ibrix_task –k –n TASKID.

You can restart the tasks again after the upgrade using the ibrix_task –r –n TASKID

command.

For more information about using the ibrix_task command, see the command help or

the HP StoreAll OS CLI Reference Guide.

Run the following command to check the overall health of the system. Even if the upgrade

script performs this health check, it is a best practice to check the system health now and

fix any issues before running the upgrade script.

ibrix_health -l

If any node does not show “PASSED” as the status, execute the following command (where

HOSTNAME is the name of the failing node) to gather more information to fix the issue:

ibrix_health –i –h <HOSTNAME>

You can also running the following commands for additional health checks of the system:6

• ibrix_vs -i (checks the storage hardware)

• ibrix_chassis -i -s (checks the server hardware and software)

• ibrix_event –l | more (displays details about events)

Execute the following command to determine if High Availability has been configured:

ibrix_haconfig –l

The upgrade will run if High Availability is configured. The purpose of this step is to verify

the state of High Availability before the upgrade. After the upgrade completes, you verify

the state of High Availability again to ensure it matches the state before the upgrade.

NOTE:

• You can run this command with the –i and –h <HOSTNAME> parameters to view more

detail, if needed.

• Ignore warnings about HBAs when running this command on the 8800.

Stop all client I/O to the cluster or file systems, so that the upgrade script does not kill these

processes and attempt to unmount the file system. On the Linux client, use lsof

<mountpoint> to show open files belonging to active processes.

To find the mount points for a file system:

• CLI: ibrix_mountpoint -l

• StoreAll Management Console: Select Filesystems in the Navigator pane.

Completing upgrade prerequisites 7