Posted: Mon Jan 06, 2020 9:34 am Post subject: multi-instance QM: system invoked failover takes longer
Chevalier
Joined: 31 Mar 2006 Posts: 411 Location: New York
Greetings.
In a multi-instance configuration, system failure triggered fail-over always takes much longer than manual fail-over triggered by endmqm -s.
The former typically takes minutes, while the latter just a couple of seconds.
Why is it so? If I turn off the power switch off the active instance server, or disconnect the network cable, wouldn't the locks be released right away and get held by the next poll from the standby instance?
thanks for sharing any insight. _________________ pcelari
-----------------------------------------
- a master of always being a newbie
A controlled failover will explicitly drop the file locks used by the active QMgr, while the abrupt termination of the server hosting the QMgr (or breaking the network links to the NFS server) will depend opon the NFS V4 lock leases expiring.
Adjusting the NFS lock lease time downwards is likely to improve matters.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum