[Pacemaker] Error with pacemaker-1.0.6 and corosync-1.1.2:

[Brian Ferris]

I don't see any corosync-1.2.0 rpms up at clusterlab.org rpm site.  It looks
like corosync-1.2.0 has only been out for six days?  Would the
pacemaker-1.0.6 from Nov 04 really be built against it?


On Mon, Dec 14, 2009 at 1:20 AM, Andrew Beekhof <andrew at beekhof.net> wrote:

> On Fri, Dec 11, 2009 at 5:34 PM, Brian Ferris <bdferris at gmail.com> wrote:
> > I checked to verify that there were no other openais or corosync
> processes
> > running.
> >
> > I think that crm_signal_dispatch signal termination message you saw was
> me
> > manually killing the corosync process tree, or at least attempting to
> stop
> > things.  I've attached a new trace that shows me running corosync for a
> good
> > six minutes before attempting to kill it.  You still see the
> > crm_signal_dispatch about six minutes in when I'm attempting to shut
> things
> > down.  I'm not sure it's a symptom of the actual underlying problem (of
> > course, I could be wrong).
> >
> > Any other ideas?  Anything additional I could enable in terms of logging
> /
> > debugging that might show useful diagnostic info?
>
> perhaps adding "debug: on" in corosync.conf, but I see:
>
> Dec 11 08:22:49 localhost corosync[1632]:   [MAIN  ] Corosync Cluster
> Engine ('1.1.2'): started and ready to provide service.
>
> IIRC, the packages were built against 1.2.0... maybe thats causing the
> problem.
> Check for a corosync update?
>
> >
> > Thanks,
> > Brian
> >
> >
> > On Fri, Dec 11, 2009 at 2:13 AM, Andrew Beekhof <andrew at beekhof.net>
> wrote:
> >>
> >> Could you check the ps axf output for a second corosync or openais
> >> process?
> >> Something out there is sending SIGTERM to the pacemaker daemons:
> >>
> >> Dec 10 11:48:12 localhost pengine: [23775]: info: crm_signal_dispatch:
> >> Invoking handler for signal 15: Terminated
> >>
> >> On Thu, Dec 10, 2009 at 9:02 PM, Brian Ferris <bdferris at gmail.com>
> wrote:
> >> > I'm running pacemaker-1.0.6 and corosync-1.1.2 on Fedora 11 using rpms
> >> > available from clusterlabs.org.
> >> >
> >> > I'm new to this stuff, and I've been having some trouble getting
> things
> >> > started.  Specifically, I'm seeing the following messages in my logs
> >> > when I
> >> > attempt to startup corosync:
> >> >
> >> > Dec 10 11:47:47 localhost attrd: [23774]: info: init_ais_connection:
> >> > Creating connection to our AIS plugin
> >> > Dec 10 11:47:47 localhost attrd: [23774]: info: init_ais_connection:
> >> > Connection to our AIS plugin (9) failed: Try again (6)
> >> >
> >> > I see that same sequence of messages for attrd, cib, stonithd.  They
> are
> >> > repeated endlessly until the programs decide to stop retrying.  This
> >> > seems
> >> > to be the closest thing I can find to a root error, but I'm not sure
> how
> >> > to
> >> > address this.
> >> >
> >> > I've attached my full message log, my corosync.conf, and cib.xml file,
> >> > if
> >> > any of that helps.
> >> >
> >> > Any ideas?
> >> >
> >> > Thanks,
> >> > Brian
> >> >
> >> > _______________________________________________
> >> > Pacemaker mailing list
> >> > Pacemaker at oss.clusterlabs.org
> >> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> >> >
> >> >
> >>
> >> _______________________________________________
> >> Pacemaker mailing list
> >> Pacemaker at oss.clusterlabs.org
> >> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> >
> >
> > _______________________________________________
> > Pacemaker mailing list
> > Pacemaker at oss.clusterlabs.org
> > http://oss.clusterlabs.org/mailman/listinfo/pacemaker
> >
> >
>
> _______________________________________________
> Pacemaker mailing list
> Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.clusterlabs.org/pipermail/pacemaker/attachments/20091214/888d75eb/attachment-0001.html>