Nikola Ciprich wrote:
Hi All,
I'd like to ask for help with following matter:
I have 2 node cluster, on top of which I have multiple cloned resources, always
2 clones, 1 clone per node.
Everything works as expected, but there is one problem:
If one of nodes is stopped (or put into standby or whatever), and then started
again, resource on OTHER node always gets restarted for some reason. In log, I
then see something like this:
May 3 20:17:52 vbox3 pengine: [4704]: notice: NoRoleChange: Move resource
clvmd:0 (vbox3 -> vbox4)
May 3 20:17:52 vbox3 pengine: [4704]: notice: StopRsc: vbox3 Stop clvmd:0
May 3 20:17:52 vbox3 pengine: [4704]: notice: StartRsc: vbox4 Start clvmd:0
May 3 20:17:52 vbox3 pengine: [4704]: notice: StartRsc: vbox3 Start clvmd:1
It seems like HB tries to "swap" the resources for some reason.
I guess problem might be in some configuration error which is causing following
warning I'm getting:
WARN: native_assign_node: 2 nodes with equal score (+INFINITY) for running the
listed resources (chose vbox4)
That probably means I should set prefered location for each of the clones, but
I'm not sure on how to do that.
I tried creating constraints for clone resources themselv as well as clone
"parts" (ie clvmd:0), but none of this helped.
Could somebody more skilled give me a hint on where the problem could be??
Any help would be greatly appreciated
thanks a lot in advance!
BR
nik
Which version do you use? Beekhof did some work on shuffling clone
instances lately.
Your config and logs would help, too.
Regards
Dominik
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
