Bug 1324022

Summary:	RFE: monitoring pacemaker cluster with nagios
Product:	Red Hat Enterprise Linux 7	Reporter:	nikhil kshirsagar <nkshirsa>
Component:	pacemaker	Assignee:	Ken Gaillot <kgaillot>
Status:	CLOSED NOTABUG	QA Contact:	cluster-qe <cluster-qe>
Severity:	low	Docs Contact:
Priority:	unspecified
Version:	7.2	CC:	abeekhof, cluster-maint, rscarazz
Target Milestone:	rc	Keywords:	FutureFeature
Target Release:	---
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:		Doc Type:	Enhancement
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2016-04-20 11:48:33 UTC	Type:	Bug
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:

Description nikhil kshirsagar 2016-04-05 11:13:39 UTC

Description of problem:
A customer wants to know if we have any plugins to support monitoring pacemaker cluster with nagios. We have a kcs https://access.redhat.com/solutions/2020263 related to this, but this is not what the customer wants to do. he does not want to add nagios as a pacemaker resource. He wants to monitor pacemaker itself with nagios, using a plugin.

Case 01608832 in SFDC.


Steps to Reproduce:
N/A

Actual results:
N/A


Expected results:
N/A


Additional info:


-------- Forwarded Message --------
Subject: 	Re: Regarding bz https://bugzilla.redhat.com/show_bug.cgi?id=1212632
Date: 	Mon, 4 Apr 2016 17:11:54 -0500
From: 	Chris Feist <cfeist>
To: 	Fabio M. Di Nitto <fdinitto>
CC: 	nkshirsa, Cole Towsley <ctowsley>


Nikhil,

It sounds like the customer is asking for a nagios monitoring agent
for pacemaker, so they can monitor (with nagios) if pacemaker is up
and running properly on a cluster.

The bug that you referred to is talking about running nagios under
pacemaker (a nagios resource agent).  We are planning to have that
implemented in an upcoming release, but I don't believe we currently
have any nagios plugins/monitoring agents that monitor pacemaker.
There are a few ones available in the community but they aren't
supported by RedHat.

The best thing to do to get this in RHEL is to file a bugzilla against
pacemaker requesting a Nagios agent, then we can evaluate and track
the request to see what we can do.

Thanks,
Chris


>>
>> A customer who has opened a case with GSS asks
>>
>> "
>> I would like to monitor our cluster RHEL7 pacemaker with Nagios.
>> Are there some plugins to do this?
>> "
>>
>> I saw your comments on bz 1212632 , and wanted to know whether we are
>> supporting what the customer wants in RHEL 7.3 / 7.2 ?
>>
>> Any info on this would be hugely helpful and much appreciated !
>>
>> Regards,
>> nikhil.

Comment 2 Andrew Beekhof 2016-04-06 01:38:47 UTC

We don't have such a plugin but one could be created based on either the 'pcs status' or 'pcs status xml' output

Comment 3 Raoul Scarazzini 2016-04-06 05:55:38 UTC

Why not using crm_mon (which is part of the package pacemaker-cli) via nrpe and obtaining something like this:

[heat-admin@overcloud-controller-0 ~]$ sudo crm_mon -s
CLUSTER OK: 3 nodes online, 112 resources configured

That's what the -s option was meant for. Of course is not pcs, but maybe it's good enough.

Comment 8 Ken Gaillot 2016-04-06 14:01:11 UTC

In the past, Red Hat did not provide or support nagios itself, so we could not provide or support any nagios plugins either, but I am trying to clarify whether that is still the case, and will comment once I find out the details.

As mentioned in Comment 3, there is already a "monitoring plugin mode" for our supported crm_mon tool. Running crm_mon with the -s or --simple-status option will produce output suitable for interpretation by nagios. It can used directly as a nagios plugin to monitor a pacemaker cluster.

Its functionality is limited:

* It will never go to nagios' critical state, only OK or warning.

* It will go to warning state if the cluster is not running, if there is no DC, or if any node is offline.

More thorough plugins exist in the nagios community. Those will be unsupported, and may need minor modifications to work in any particular user's environment, but they do provide a detailed status of a pacemaker cluster.

I will comment further once I have more information.

Comment 12 Ken Gaillot 2016-04-07 14:45:48 UTC

Red Hat does not provide or support nagios itself. A couple of projects (gluster and packstack) do provide some nagios integration capability, but these are special cases and not applied broadly. Therefore, we will not provide or support one of the cluster agents from the nagios community.

However, we do provide and support pacemaker's crm_mon tool, including its -s/--simple-status option, which can be used directly as a nagios plugin. So, that is our supported solution for this situation.

If the current capabilities of crm_mon -s are not sufficient for the customer's purposes, we can use this BZ as a request to enhance it.