Bug 1751142

Summary:	[downstream clone - 4.3.6] host activation causes RHHI nodes to lose the quorum
Product:	Red Hat Enterprise Virtualization Manager	Reporter:	RHV bug bot <rhv-bugzilla-bot>
Component:	ovirt-engine	Assignee:	Gobinda Das <godas>
Status:	CLOSED ERRATA	QA Contact:	SATHEESARAN <sasundar>
Severity:	urgent	Docs Contact:
Priority:	urgent
Version:	4.3.4	CC:	aoconnor, emarcus, godas, joboyer, lsantann, mkalinin, obockows, pelauter, rcarrier, rcyriac, Rhev-m-bugs, rhinduja, sabose
Target Milestone:	ovirt-4.3.6	Keywords:	ZStream
Target Release:	4.3.6
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	ovirt-engine-4.3.6.6	Doc Type:	Bug Fix
Doc Text:	During host activation, the engine did not check the glusterd status. The gluster service was restarted even though the glusterd was not stopped during maintenance. In this release, the quorum is not lost upon host activation.	Story Points:	---
Clone Of:	1741102	Environment:
Last Closed:	2019-10-10 15:36:58 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	Gluster	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:
Bug Depends On:	1741102
Bug Blocks:	1751717

Description RHV bug bot 2019-09-11 09:24:22 UTC

+++ This bug is a downstream clone. The original bug is: +++
+++   bug 1741102 +++
======================================================================

Description of problem:
When a host providing gluster services for RHHI is activated, the RHV-M initiates gluster services restart. This causes gluster volumes to lose quorum and stop working.


Version-Release number of selected component (if applicable):
rhvm 4.3.3

How reproducible:
100%

Steps to Reproduce:
1. Put a gluster node into maintenance
2. Set debug level
    vdsm-client Host setLogLevel level=DEBUG name=jsonrpc
#. Activate the node again

Actual results:
2019-08-14 09:30:00,866+0000 DEBUG (jsonrpc/1) [jsonrpc.JsonRpcServer] Calling 'GlusterService.action' in bridge with {u'action': u'restart', u'serviceNames': [u'glusterd']} (__init__:329)

Expected results:
No restart happens. Keep in mind that the node can be put into maintenance mode without stopping the gluster services.

(Originally by Roman Hodain)

Comment 4 RHV bug bot 2019-09-11 09:24:30 UTC

sync2jira

(Originally by Daniel Gur)

Comment 5 RHV bug bot 2019-09-11 09:24:31 UTC

sync2jira

(Originally by Daniel Gur)

Comment 11 SATHEESARAN 2019-09-25 16:48:32 UTC

Verified with RHV 4.3.6.6 with the following steps

1. Created a HC cluster
2. Move the HC node in to MAINTENANCE, without stopping gluster services
3. Note the PID of glusterd process and brick process.
4. Activate the host back

Observed that the gluster process PID remained the same post the host is activated

Comment 18 errata-xmlrpc 2019-10-10 15:36:58 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-2019:3010