Hello,
we would like to test a scenario of replacing a failed node in HA configuration, with replacement node configured to look like the failed node, as described here:
https://www.ibm.com/docs/en/ibm-mq/9.3?topic=availability-replacing-failed-node-in-high-configuration
Can we simulate the node failure by moving rdqm data to a backup location, as a first attempt, to see that we can restore everything from drbd (pacemaker) cluster, without bringing data back from the backup location? How exactly this should be done? What needs to be moved, what should be left in order to get everything back, is pacemaker service supposed to be stopped during tacking backup?
Thanks in advance.

Hrvoje,

Make sure you move all the qmgrs off the intended server, then change the hostname and ips on the server, shut down the server

Slide the new server in place of the shut down one. Give it the hostname and ips of the shutdown server (as they were before you changed them)

set up the rdqm.ini (copy from one of the other servers, follow the procedure.

Francois,

thank you.

Best regards