[Linux-HA] pingd cloned resource question

[Andrew Beekhof]

this bug was fixed in 2.0.8

On 3/13/07, Valentin Eberhardt <veberhardt at amadeus.com> wrote:
> Hi!
>
> In a 2-node symmetrical cluster based on 2.0.7./ES10 a resource group and
> a clone resource
> of type pingd is  configured. Two ping targets are defined in ha.cf. When
> one of the ping
> targets is not answering the pings from node-A anymore , the pingd-child:0
> goes into status
> "Stopped" and node-A is marked OFFLINE.
>
> 1. Question: Why? One ping target is still alive as seen from node-A. I
> just expected that the
>  pingd score is reduced by the multiplier value (from 2x multiplier to 1x
> multiplier). And not
> more. The daemon pingd is still running, it's not stopped.
>
> 2. Now the resource group runs on node-B. Which is OK.
> If one executes: # crm_resource -C -r pingd-child:0
>
> the resource pingd-child:0 goes back in status "Started" and node-A to
> ONLINE.
> While this happens the resource group is switched once back and forth
> instead of just
> staying where it is.
> Any idea why?
>
> Thanks.
> Val
> _______________________________________________
> Linux-HA mailing list
> Linux-HA at lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
>