Bug 210186

Summary:	qdisk man page updates
Product:	[Retired] Red Hat Cluster Suite	Reporter:	Lon Hohberger <lhh>
Component:	cman	Assignee:	Lon Hohberger <lhh>
Status:	CLOSED WONTFIX	QA Contact:	Cluster QE <mspqa-list>
Severity:	medium	Docs Contact:
Priority:	medium
Version:	4	CC:	cluster-maint, jbacik
Target Milestone:	---
Target Release:	---
Hardware:	All
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2006-11-28 20:01:01 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:

Description Lon Hohberger 2006-10-10 17:37:43 UTC

Answers to frequently asked questions about qdisk need to be included in the
qdisk(5) man page.

* Include minimum qdisk size in documentation.
* Talk about the cman deadnode_timeout and the qdisk timeout and how they
interrelate, and how to extend the openais timeout on RHEL5 / FC6 / etc.
* Include some real-world examples with failure cases, recovery, expected
behaviors, etc.

Comment 1 Josef Bacik 2006-10-10 19:28:05 UTC

* Indicate that post_join_delay should be 2x the tko time for qdisk
* For two node clusters, cman still has to have the two_node="1" and
expected_votes="1" options set.

I think thats it for now.

Comment 2 Lon Hohberger 2006-10-11 17:04:03 UTC

* Documentation concerning cman<-->qdisk interaction is weak (or really,
nonexistent!)

Comment 3 Lon Hohberger 2006-10-11 17:10:19 UTC

* I don't think two_node is needed for 2-node clusters, which would obviate
post_join_delay.  However, I should note that it may be desirable to decrease
the CMAN join timeout in /etc/init.d/cman

Comment 4 Kiersten (Kerri) Anderson 2006-11-28 19:56:13 UTC

devel NAK for 4.5 - removing from 4.5 list.  Any updates to the man pages would
be dependent on the current cman/qdisk relationship discussion that is ongoing
and would be made as part of that effort.  Lon, can we just close this one?

Comment 5 RHEL Program Management 2006-11-28 20:01:01 UTC

Development Management has reviewed and declined this request.  You may appeal
this decision by reopening this request.