Bug 1247019

Summary:	must enable fence device naming workaround
Product:	Red Hat OpenStack	Reporter:	Fabio Massimo Di Nitto <fdinitto>
Component:	rhosp-director	Assignee:	Chris Jones <chjones>
Status:	CLOSED ERRATA	QA Contact:	Asaf Hirshberg <ahirshbe>
Severity:	low	Docs Contact:
Priority:	low
Version:	Director	CC:	abeekhof, ahirshbe, chjones, cluster-maint, cluster-qe, dmacpher, fdinitto, hbrock, jruemker, lruzicka, mburns, mtanino, oblaut, rhel-osp-director-maint, royoung, srevivo, tsekiyam, ushkalim
Target Milestone:	rc	Keywords:	InstallerIntegration, TestOnly, Triaged
Target Release:	11.0 (Ocata)
Hardware:	x86_64
OS:	Unspecified
URL:	https://api.access.redhat.com/rs/telemetry/abrt/reports/bthash/c060d3a4a86c68a93275db70569ad33e071e8bc9
Whiteboard:	abrt_hash:566d6d8d0ad5cef6e984df4fa09b28d7355d8e53
Fixed In Version:		Doc Type:	Known Issue
Doc Text:	Pacemaker continuously crashes when the fencing device name and the host name are the same. To avoid this problem, add the "fence-" prefix or the "-fence" suffix to the name of the fencing device. With the names configured like this, the cluster works without errors.	Story Points:	---
Clone Of:	1246291	Environment:
Last Closed:	2017-05-17 19:23:43 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	1242422, 1246291, 1558241
Bug Blocks:

Comment 3 Fabio Massimo Di Nitto 2015-07-27 07:43:01 UTC

Skipping trough the details of the original bug, the way in which ospd is configuring fence devices causes pacemaker to crash.

The root issue is of course in pacemaker and we are working to fix it, BUT we can´t make it in time for OSP7 GA.

The problem is that OSPd is configuring fence devices with a name that matches the hostname. This makes pacemaker crash and fail.

the simplest workaround possible is to simply add a fence-* prefix or *-fence suffix when configuring stonith devices.

Without this workaround and the proper fix, it will be impossible to manage a HA OSP.

Be aware that everything will work just fine after a normal deployment because fence devices are configured after cluster has formed, but everything will explode at the first time one of the nodes is rebooted or restarted.

Comment 4 chris alfonso 2015-08-20 16:34:22 UTC

Moving this to A2 since we don't yet support fence device configuration with OSP-director just yet.

Comment 7 Mike Burns 2016-04-07 20:47:27 UTC

This bug did not make the OSP 8.0 release.  It is being deferred to OSP 10.

Comment 9 Fabio Massimo Di Nitto 2016-08-30 12:30:39 UTC

moving to OSP11 as part of auto-fencing implementation.

Chris, while the pacemaker bug has been fixed, we still want to have a fence-$nodename or $nodename-fence entry when creating stonith services.

Comment 17 Asaf Hirshberg 2017-05-01 08:33:43 UTC

Verified.

Comment 20 errata-xmlrpc 2017-05-17 19:23:43 UTC

Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHEA-2017:1245