Does it do this before or after the node starts coming up again?
On Dec 5, 2008, at 9:57 AM, Satomi TANIGUCHI wrote:
Hi,
I found an odd behavior which appears after startup-fencing.
After startup-fencing succeeded, I stopped Heartbeat service.
Then, certainly it stopped, but it tried to STONITH again and again
in stop processing.
When this behavior appears, HB shows following messages.
　crmd[18679]: 2008/12/05_17:01:26 ERROR: cib_fencing_updated:
CIB update
failed: The object/attribute does not exist
After STONITH succeeds, CIB tries to delete <lrm/> and
<transient_attributes/>
in <node_status/> of STONITH'ed node.
But in the case of startup-fencing,
it doesn't have target node's <node_status/> yet.
Maybe it causes this bug, I think.
Please see the attached hb_report's archive for details.
Best Regards,
Satomi TANIGUCHI
<hb_report.tar.bz2>_______________________________________________
Pacemaker mailing list
[email protected]
http://list.clusterlabs.org/mailman/listinfo/pacemaker
_______________________________________________
Pacemaker mailing list
[email protected]
http://list.clusterlabs.org/mailman/listinfo/pacemaker
