[Pacemaker] cman + corosync + pacemaker + fence_scsi

[Andrew Beekhof]

On 26/04/2013, at 4:25 PM, Angel L. Mateo <amateo at um.es> wrote:

> El 26/04/13 02:01, Andrew Beekhof escribió:
>> 
>> On 24/04/2013, at 10:48 PM, Angel L. Mateo <amateo at um.es> wrote:
>> 
>>> Hello,
>>> 
>>> 	I'm trying to configure a 2 node cluster in ubuntu with cman + corosync + pacemaker (the use of cman is because it is recommended at pacemaker quickstart). In order to solve the split brain in the 2 node cluster I'm using qdisk.
>> 
>> If you want to use qdisk, then you need something newer than 1.1.8 (which did not know how to filter qdisk from the membership).
>> 
> 	Oopps. I have cman 3.1.7, corosync 1.4.2 and pacemaker 1.1.6 (the ones provided with ubuntu 12.04).
> 
> 	My purpose for using qdisk is to solve split brain problem in my two nodes cluster. Another suggestion for this?

Another node (with standby=true) might be an option.
Or try and get a newer version.

> 
>> 
>>> For fencing, I'm trying to use fence_scsi and in this point I'm having the problem. I have attached my cluster.conf.
>>> 
>>> xml <node id="/dev/block/8:33" type="normal" uname="/dev/block/8:33"/>
>>> node myotis51
>>> node myotis52
>>> primitive cluster_ip ocf:heartbeat:IPaddr2 \
>>> 	params ip="155.54.211.167" \
>>> 	op monitor interval="30s"
>>> property $id="cib-bootstrap-options" \
>>> 	dc-version="1.1.6-9971ebba4494012a93c03b40a2c58ec0eb60f50c" \
>>> 	cluster-infrastructure="cman" \
>>> 	stonith-enabled="false" \
>>> 	last-lrm-refresh="1366803979"
>>> 
>>> 	At this moment I'm trying just with an IP resource, but at the end I'll get LVM resources and dovecot server running in top of them.
>>> 
>>> 	The problem I have is that whenever I interrupt network traffic between my nodes (to check if quorum and fencing is working) the IP resource is started in both nodes of the cluster.
>> 
>> Do both side claim to have quorum?
>> Also, had you enabled fencing the cluster would have shot its peer before trying to start the IP.
>> 
> 	I think I did (and this configuration with stonith disabled is because modified for later tests) but I will check it again.
> 
>>> 
>>> 	So it seems that node fencing configure at cluster.conf is not working for me.
>> 
>> Because pacemaker cannot use it from there.
>> You need to follow
>> 
>>    http://clusterlabs.org/doc/en-US/Pacemaker/1.1-plugin/html/Clusters_from_Scratch/_configuring_cman_fencing.html
>> 
>> and then teach pacemaker about fence_scsi:
>> 
>>    http://clusterlabs.org/doc/en-US/Pacemaker/1.1-plugin/html/Clusters_from_Scratch/ch09.html
>> 
>>> Then I have tried to configure as a stonith resource (since it is listed by sudo crm ra list stonith), so I have tried to include
>>> 
>>> primitive stonith_fence_scsi stonith:redhat/fence_scsi
>>> 
>>> 	The problem I'm having with this is that I don't know how to indicate params for the resource (I have tried params devices="...", params -d ..., but they are not accepted) and with this (default) configuration I get:
>> 
>> See the above link to chapter 9.
>> 
> 	I have tried this. The problem I'm having is that I don't know how to create the resource using fence_scsi. I have tried different syntaxes
> 
> crm(live)configure# primitive stonith_fence_scsi stonith:redhat/fence_scsi \

Remove the "redhat/" part. 

> > params name="scsi_fence" devices="/dev/sdc"
> ERROR: stonith_fence_scsi: parameter name does not exist

'name' is the name of the machine to be shot and is filled in at runtime 

> ERROR: stonith_fence_scsi: parameter devices does not exist

This looks like crmsh not knowing how to find the agent's metadata and can be ignored.

"man fence_scsi" looks like you need a value for "key" though

> 
> crm(live)configure# primitive stonith_fence_scsi stonith:redhat/fence_scsi \
> > params n="scsi_fence" d="/dev/sdc"
> ERROR: stonith_fence_scsi: parameter d does not exist
> ERROR: stonith_fence_scsi: parameter n does not exist
> 
> crm(live)configure# primitive stonith_fence_scsi stonith:redhat/fence_scsi \
> > params -n="scsi_fence" -d="/dev/sdc"
> ERROR: stonith_fence_scsi: parameter -d does not exist
> ERROR: stonith_fence_scsi: parameter -n does not exist
> 
> 	Does anyone has an example for this? What I would like to do is that in case of problems, the node with the use of scsi channel (the one using my LMV volumes) shoots the other one. Could I use the same behaviour with external/sbd stonith resource?
> 
> -- 
> Angel L. Mateo Martínez
> Sección de Telemática
> Área de Tecnologías de la Información
> y las Comunicaciones Aplicadas (ATICA)
> http://www.um.es/atica
> Tfo: 868889150
> Fax: 868888337
> 
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> 
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs: http://bugs.clusterlabs.org