Am 04.04.2011 10:32, schrieb Andrew Beekhof:
> On Thu, Mar 24, 2011 at 9:58 PM, Christoph Bartoschek
> <[email protected]>  wrote:
>> It seems as if the g_nfs service is stopped on the surviving node when
>> the other one comes up again.
>
> To me it looks like the service gets stopped after it fails:
>
>      p_exportfs_root:0_monitor_30000 (node=laplace, call=12, rc=7,
> status=complete): not running
>

Hi,

thanks for the answer. In the meantime I think that the problem is a 
result of two separate issues:

1. It is wrong to have ocf::heartbeat:exportfs running separate from the 
group g_nfs.  It has to start after the filesystem comes up and before 
the ip is assigned. Otherwise it will fail during restart in some 
situations.

2. The cluster tries to restart the service on the surviving node after 
the crashed one comes up.

Christoph
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to