On Oct 21, 2009, at 5:44 AM, Paolo Pisati wrote:

> Paolo Pisati wrote:
>> Dear guys,
>>
>> i've a small problem with an NFS/drbd/heartbeat cluster: basically  
>> the
>> secondary node (that was previously promoted
>> as primary) when the primary come up again, is unable to release the
>> resources (ip/drbd) gracefully, and reboots.
>> I know there's resource stickiness with pacemaker, but i could find  
>> any
>> info for plain heartbeat.
>> Moreover, is the behavior of node2 normal in my case? (config files  
>> are
>> below).
>>
> Never mind: the culprit was the bloody nfs centos init.d script that,
> actually, didn't kill nfsd, that in turns
> didn't let /etc/ha.d/resource.d/Filesystem stop/umount the filesystem,
> etcetc
>
> changing /etc/init.d/nfs from:
>
> stop)
> # Stop daemons.
> echo -n $"Shutting down NFS mountd: "
> killproc rpc.mountd
> echo
> echo -n $"Shutting down NFS daemon: "
> killproc nfsd -2
> echo
>
> to:
>
> stop)
> # Stop daemons.
> echo -n $"Shutting down NFS mountd: "
> killproc rpc.mountd
> echo
> echo -n $"Shutting down NFS daemon: "
> killproc nfsd -KILL
> echo
>
> solved my problem.
>
> (see here:
> http://www.centos.org/modules/newbb/viewtopic.php?viewmode=flat&topic_id=11739&forum=41)
>  
> .
> i think it's worth an entry in the FAQ IMHO.

This is mentioned in the HaNFS tutorial.  See #3 in the 'Hints' section.

http://www.linux-ha.org/HaNFS

>
>
> now, if i can get the ip not to bounce around when nodes are  
> resurrected
> (aka resource stickiness), that would be perfect...
> _______________________________________________
> Linux-HA mailing list
> [email protected]
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems

_______________________________________________
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