[Pacemaker] How to recover from an unclean shutdown

Andrew Beekhof andrew at beekhof.net
Tue Nov 1 06:33:56 EDT 2011

We discussed this on IRC yesterday didnt we?
You're trying to stop drbd while apache is still running (group
ordering it wrong) which is likely to cause both to fail and prevent
the cluster form terminating.

On Mon, Oct 31, 2011 at 4:21 AM, Mailing List SVR
<lists at svrinformatica.it> wrote:
> Hi,
> I'm a long time heartbeat (v1) users, I'm now experimenting with pacemaker
> and corosync using two virtual machine (ubuntu oneiric, corosync and
> pacemaker from repo) to prepare the migration of my clusters.
> After configuring some services I uncleany shutdown a virtual machine
> (server1 was hunging on stopping corosync since I did a configuration
> mistake) now only the service that was on server2 start, I think the problem
> is the following:
> Oct 30 17:24:47 server2 pengine: [2323]: debug: native_assign_node: All
> nodes for resource drbd_disk:1 are unavailable, unclean or shutting down
> (server1: 0, -1000000)
> how recover from this situation? (I already tryed crm node clearstate
> server1)
> Please note that if I stop corosync and I manually start drbd the drbd
> resources starts fine.
> Attached are the whole logs and my conf, I can share the virtual machines
> too if needed,
> thanks
> Nicola
> _______________________________________________
> Pacemaker mailing list: Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> Project Home: http://www.clusterlabs.org
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf
> Bugs:
> http://developerbugs.linux-foundation.org/enter_bug.cgi?product=Pacemaker

More information about the Pacemaker mailing list