CTDB doesn't start

[Tim Serong]

On 08/07/11 21:11, Андрей Гребенников wrote:
> Hi!
> I use the pacemaker as a clusterware, and I want to stop one node. After
> that comes the next:
>
> 2011/07/08 15:09:56.977465 [30457]: Recmaster node 0 no longer
> available. Force reelection
> 2011/07/08 15:09:59.980806 [30457]: server/ctdb_recoverd.c:905 Election
> timed out
>
> and these messages don't end to appear. Then the CTDB comes stopped.
> What's that?

You mean you shutdown corosync/openais/pacemaker on one node, and CTDB 
died on the other node?  It might help if you can post a copy of your 
pacemaker config (output of "crm configure show"), and any other 
relevant configuration.

Regards,

Tim
-- 
Tim Serong <tserong at novell.com>
Senior Clustering Engineer, OPS Engineering, Novell Inc.