Mailing List Archive: errors with heartbeat/drbd

Christian Guertler wrote:
>
> Hello;
>
> we are using heartbeat and drbd. When heartbeat on node1 is shut down, the
> takeover by the 2nd node works fine. But when heartbeat on the 1st node is
> started again, sometimes the session on the second node is closed and
> following messages occur in /var/log/messages:
>
> Jan 24 14:54:56 ra heartbeat: info: Running /etc/ha.d/resource.d/datadisk
> drbd0
> stop
> Jan 24 14:54:56 ra datadisk: failed
> Jan 24 14:54:57 ra PAM_pwdb[2877]: (su) session closed for user root
> Jan 24 14:55:04 ra datadisk: succeeded
> Jan 24 14:55:04 ra datadisk: succeeded
>
> But the processes are working ok.
> Is there a false configuration?

Looks like a heartbeat configuration/setup problem. When heartbeat is
restarted on node1 it should either always or never stop the services on
the second node (depending on the configuration -> nice_fallback). Doing
so _sometimes_ is weird.

Greetings,
Juri

--
juri.haberland@example.com
system engineer innominate AG
clustering & security the linux architects
tel: +49-30-308806-45 fax: -77 http://www.innominate.com