Hi all,
There is the configuration for my Xen ressource.
<primitive id="qclvmsles01" class="ocf" type="Xen" provider="heartbeat">
<operations>
<op name="monitor" interval="10s" timeout="30s" id="xen-op-01"
start_delay="0" disabled="false" role="Started"/>
<op name="stop" timeout="30s" id="xen-op-02" start_delay="0"
disabled="false" role="Started"/>
</operations>
<instance_attributes id="qclvmsles01_instance">
<attributes>
<nvpair id="xen-01" name="xmfile"
value="/etc/xen/vm/qclvmsles01"/>
<nvpair id="qclvmsles01_target_role" name="target_role"
value="started"/>
</attributes>
</instance_attributes>
<meta_attributes id="qclvmsles01_meta">
<attributes>
<nvpair id="xen-02" name="allow_migrate" value="true"/>
</attributes>
</meta_attributes>
</primitive>
I have a location constraints forcing the VM to go back on node1 when it's
possible.
so :
node1 (vm01)
node2
----
node1 goes offline
vm01 start on node2
when node1 comeback online the vm01 start to live migrate from node2 to
node1. During the live migration the RA seems to lost the status of the
VM... The ressource just goes in "Not Running" status but it still
migrating.. and if I wait few second the migration complete sucesfully..
Then I can clean the ressource and the RA will see where the vm is
running...
I tried to ajust the timeout delay ... but if I raise it when node1 crash it
take long time before the vM start on node2...
Is there another king of timeout I can set ? someting like migration timeout
?
Thanks
--
René Jr Purcell
Chargé de projet, sécurité et sytèmes
Techno Centre Logiciels Libres, http://www.tc2l.ca/
Téléphone : (418) 681-2929 #124
_______________________________________________
Linux-HA mailing list
Linux-HA@lists.linux-ha.org
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
