There is a bug in the Xen resource agent. monitor calls fail if monitor is called during migrations. To trigger det bug - just set up HB2 to use live migrations and flip one of the nodes into and out of standby mode. After a short while HB2 will get the "resource running on two nodes"-error.
The following patch should take care of it. vm's are in state "1" when they are being migrated. Per.
--- /root/hb2/heartbeat-2.1.2/debian/heartbeat/usr/lib/ocf/resource.d/heartbeat/Xen 2007-08-25 18:20:27.000000000 +0200
+++ /usr/lib/ocf/resource.d/heartbeat/Xen 2007-08-26 20:14:25.000000000 +0200
@@ -80,7 +82,7 @@
Xen_Status() {
STATUS=`xm list --long $DOMAIN_NAME | grep status 2>/dev/null`
STATUS_NOSPACES=`echo "$STATUS" | awk '{ print $1,$2}'`
- if [ "$STATUS_NOSPACES" = "(status 2)" ] ; then
+ if [ "$STATUS_NOSPACES" = "(status 2)" -o "$STATUS_NOSPACES" = "(status 1)" ] ; then
return $OCF_SUCCESS
else
return $OCF_NOT_RUNNING
signature.asc
Description: OpenPGP digital signature
_______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
