Bug 693427

Summary:	fence_ipmilan returns incorrect status on monitor op if chassis is powered off
Product:	Red Hat Enterprise Linux 5	Reporter:	Marek Grac <mgrac>
Component:	cman	Assignee:	Marek Grac <mgrac>
Status:	CLOSED ERRATA	QA Contact:	Cluster QE <mspqa-list>
Severity:	high	Docs Contact:
Priority:	high
Version:	5.6	CC:	bubble, cluster-maint, djansa, edamato, fdinitto
Target Milestone:	rc
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	cman-2.0.115-80.el5	Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:	690735	Environment:
Last Closed:	2011-07-21 07:26:25 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Bug Depends On:	690735, 693428
Bug Blocks:	693426

Description Marek Grac 2011-04-04 15:57:46 UTC

+++ This bug was initially created as a clone of Bug #690735 +++

Description of problem:
There is a bug in fence_ipmilan agent: it exits with result code 2 on "monitor"/"status" command if chassis is powered off. Instead it should return 0 because "monitor" command should return fence device status, not chassis status.

Version-Release number of selected component (if applicable):
fence-agents-3.1.2-1.fc13.x86_64

--- Additional comment from bubble on 2011-03-25 04:24:11 EDT ---

Created attachment 487504 [details]
Patch to fix issue

--- Additional comment from mgrac on 2011-03-25 06:32:16 EDT ---

Created attachment 487518 [details]
Proposed patch

@Vladislav:

Thanks for reporting this bug. You are right, we should return 0 if the fence device is working. 

I have created a patch that should fix this. Yours patch will work correctly for 'monitor' but it will break 'status' as fence agent won't be able to return that chassis is powered off.

--- Additional comment from bubble on 2011-03-25 08:11:02 EDT ---

Looking through the code, it was probably just forgotten (ret is set to 0 for both on and off states but not used then).

I agree, your patch is cleaner.

Comment 4 Dean Jansa 2011-05-11 22:45:04 UTC

[root@marathon-04 ~]# fence_ipmilan -a z2-ipmi -l XXX -p XXX -o off
Powering off machine @ IPMI:z2-ipmi...Done

[root@marathon-04 ~]# fence_ipmilan -a z2-ipmi -l XXX -p XXX -o monitor
Getting status of IPMI:z2-ipmi...Done
[root@marathon-04 ~]# echo $?
0


verified in cman-2.0.115-82.el5

Comment 5 errata-xmlrpc 2011-07-21 07:26:25 UTC

An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHBA-2011-1001.html