Upgrading RDQM (replicated data queue managers)

We can upgrade RDQM when applying a fixpack by using the supplied upgrade script.

About this task

An upgrade script is supplied with the fixpack image at MQ_FixpackDriver/Advanced/RDQM/. We must run the script as root:

updateRDQMsupport

This script upgrades the latest kmod-DRBD packages and latest pacemaker packages. kmod-DRBD is the kernel module for DRBD. If you apply an OS upgrade such that the version of the Linux kernel that is being used changes, we must upgrade the DRBD kernel module to match the new version by running updateRDQMsupport.

The script does not update IBM MQ .

We can run your queue managers while upgrading by failing over to the other nodes in the RDQM configuration in turn and upgrading one at a time. For advice on the order in which to update HA RDQM nodes, see Migrating HA RDQMs, for advice on the order in which to update DR RDQM nodes, see Migrating DR RDQMs, for advice on the order in which to update DR/HA RDQM nodes, see Migrating DR/HA RDQMs.

Procedure

• To upgrade HA RDQM support, on each node:
  1. Suspend the HA group. To suspend the HA group on the node, enter the following command:

     rdqmadm -s

  2. Run the script updateRDQMsupport to upgrade RDQM support, DRBD, and Pacemaker.
  3. Resume the HA group. To resume the HA group on the node, enter the following command:

     rdqmadm -r

• To upgrade DR RDQM support:
  1. Upgrade the recovery node by running the updateRDQMsupport to upgrade RDQM support and DRBD.
  2. Stop the queue manager running on the primary node using the endmqm command. (If required, start the queue manager on the recovery node.)
  3. Run the script updateRDQMsupport to upgrade RDQM support and DRBD.
  4. If you started the queue manager on the recovery node, end it.
  5. Start the queue manager on the primary node by using the strmqm command.

• To upgrade DR/HA RDQM support:
  1. On each node of the HA group on the recovery site, complete the following steps:
     1. Suspend the HA group. To suspend the HA group on the node, enter the following command:

        rdqmadm -s

     2. Run the script updateRDQMsupport to upgrade RDQM support, DRBD, and Pacemaker.
     3. Resume the HA group. To resume the HA group on the node, enter the following command:

        rdqmadm -r

  2. Stop the DR/HA RDQMs on the main site, or perform a managed failover to the recovery site.
  3. On each node of the HA group on the main site, complete the following steps:
     1. Suspend the HA group. To suspend the HA group on the node, enter the following command:

        rdqmadm -s

     2. Run the script updateRDQMsupport to upgrade RDQM support, DRBD, and Pacemaker.
     3. Resume the HA group. To resume the HA group on the node, enter the following command:

        rdqmadm -r

  4. Restart the DR/HA RDQMs, or fail them back to the main site.

Parent topic: Applying upgrades and fixes to IBM MQ

Related information

• rdqmadm (administer replicated data queue manager cluster)
• endmqm (end queue manager)
• strmqm (start queue manager)