[Pacemaker] Different behaviour for cloned resource on 1 and 2 nodes

[Johan Huysmans]

Hi,

I created a bug (http://bugs.clusterlabs.org/show_bug.cgi?id=5170) 
containing extra explanation and the crm_report.

Thanks for taking your time to check this problem!

gr.
Johan

On 15-07-13 04:11, Andrew Beekhof wrote:
> On 11/07/2013, at 12:21 AM, Johan Huysmans <johan.huysmans at inuits.be> wrote:
>
>> Hi All,
>>
>> I have a setup with a cloned resource and a resource group.
>> I also configured some colocation and order rules in such way that the group can only run where the cloned resource is running.
>>
>> On a 2 node setup I have no problems. The group moves away when the local clone resource fails.
>> The failing clone resource appears as unmanaged as I configured it as on-fail="block"
>>
>> When one of these 2 nodes is offline I see a different behaviour.
>> When a local clone resource fails, the group is restarted and the clone resource is restarted.
>>
>> It seems that in a 1 node setup the on-fail="block" is not followed and it performs the on-fail="restart".
>>
>> Is this normal behaviour?
> No. That doesn't seem right.
> Can you file a bug for that please?  Include a crm_report for the period covered by the test above.
>
>
> _______________________________________________
> 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