[Linux-cluster] service state unchanged when host crashes

[Lon Hohberger]

On Wed, 2009-10-28 at 10:24 +0000, Martin Waite wrote:
> Hi, 
> 
> This does look like the same problem:
> 
> martin at clusternode27:~$ sudo /usr/sbin/cman_tool -f nodes
> Node  Sts   Inc   Joined               Name
>   27   M     44   2009-10-27 14:59:33  clusternode27
>   28   M     52   2009-10-27 14:59:49  clusternode28
>   30   X     64                        clusternode30
>        Node has not been fenced since it went down
> 
> Even though the message in syslog claims that the fencing succeeded.

Your logs said fencing failed?

Oct 26 18:30:02 clusternode27 fenced[12749]: fencing node
"clusternode30"
Oct 26 18:30:02 clusternode27 fenced[12749]: fence "clusternode30"
failed

You can run:

  fence_ack_manual -e -n clusternode30

...to override failed fencing.

-- Lon