Hello Danny,
> are there any other commands or tools besides of crm_standby which can
> be used for HA cluster node maintenance? I wonder how you can
> shutdown all the nodes of your cluster properly without initiating
> failovers. What do you recommend to do in such case? I could run
> crm_standby on all nodes at the same time, but in fact they don't run
> exactly at the same time. And I can only do it for a certain node, not
> for the whole cluster. Or is it cluster-wide possible somehow?
you could set the resources to ,,stopped'':
crm_resource -r ms-drbd0 -v 'stopped' --meta -p target_role
for each resource and than call ,,/etc/init.d/heartbeat stop'' on each
node. There must be also a way to tell the cluster to stop everything by
changing the default policy. But I am to tired to look it up.
or something like this. However for me, I don't want to stop my cluster,
I want to take one node offline for updates. Than the other. Normally I
do that through ,,/etc/init.d/heartbeat stop''.
Or you could wipe the cib using the following command:
# Wipe old content (Attention don't do that during production it gets
# your service down: cibadmin -E)
I assume you have backed up your cib somewhere that you can load it
again with:
cibadmin -U -x /path/to/cib.xml
I shut down my test clusters normally by calling ,,/etc/init.d/heartbeat
stop'' on each node and wait.
Thomas
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
