[Pacemaker] Configuring a cluster for asymmetric operation

[David Vossel]

----- Original Message -----
> From: "Brad Jones" <brad at jones.name>
> To: "David Coulson" <david at davidcoulson.net>
> Cc: "The Pacemaker cluster resource manager" <pacemaker at oss.clusterlabs.org>
> Sent: Friday, June 8, 2012 10:21:51 AM
> Subject: Re: [Pacemaker] Configuring a cluster for asymmetric operation
> 
> David - thanks much.  I installed drbd, kept the skeleton (basically
> empty) config file in place on server C, and it has stopped trying to
> monitor.  I still don't quite understand how a server not running a
> particular RA can monitor a service anyway?

The server not running the RA is not monitoring the active resource on another server. The monitor action is being used to verify that the RA is _NOT_ running on the server it shouldn't be running on.  This is done when pacemaker starts up so it can be certain it knows the state of the cluster.

For example, say a daemon pacemaker is supposed to have control over was started before pacemaker... Pacemaker uses the monitor operation to detect that daemon is already running, otherwise pacemaker wouldn't know it was up and might try to start it somewhere else.

-- Vossel


> 
> But my immediate problem is indeed solved and I can keep symmetric
> "on."
> --
> Brad Jones
> Brad at Jones.name
> Mobile: 303-219-0795
> 
> 
> On Fri, Jun 8, 2012 at 4:45 AM, David Coulson
> <david at davidcoulson.net> wrote:
> > Pacemaker needs to be able to monitor on all nodes. Maybe if you
> > install
> > drbd on the third node but don't configure anything monitor will
> > correctly
> > report it is not running over there, and your location rules will
> > stop it
> > from even trying.
> >
> > Or just change the RA for DRBD to report not running instead of not
> > installed - I had to do that in a few cases where the RA needed the
> > config
> > file to exist, but the file was on a filesystem managed by
> > pacemaker. Seems
> > to work ok, and isn't impacting the cluster at all.
> >
> >
> > On 6/8/12 3:36 AM, Brad Jones wrote:
> >>
> >> I have a cluster with three nodes, A B and C.  A and B can
> >> basically
> >> stand in for one another; they run a DRBD master/slave set, and
> >> failure of A where most things run normally fires up services on
> >> B.
> >>
> >> C helps arbitrate quorum and runs Stonith plugins.  I've written
> >> location rules to prohibit most of the production services from
> >> running on node C.
> >>
> >> Problem is, the DRBD RA tries to monitor on node C, even though
> >> DRBD
> >> isn't even installed.  (It fails, helpfully, "not installed.")
> >>  I'm
> >> stuck with a slowly-incrementing failcount on C.
> >>
> >> When I give all services a default location rule and set
> >> symmetric-cluster="false", some of the resources stay running on A
> >> but
> >> most of the important ones stop.
> >>
> >> This was raised back in 2010 here:
> >> <http://oss.clusterlabs.org/pipermail/pacemaker/2010-April/005873.html>
> >> but I can't find much more about this issue on the web and there's
> >> only passing mention of it in the Pacemaker documentation.
> >>
> >> Bottom line: What's required to correctly configure a cluster with
> >> symmetric-cluster set to false?
> >>
> >> FYI I'm on pacemaker 1.1.6, heartbeat 3.0.5, on Ubuntu 10.04.
> >>
> >> Thanks!  Brad
> >> --
> >> Brad Jones
> >> Brad at Jones.name
> >> Mobile: 303-219-0795
> >>
> >> _______________________________________________
> >> 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://bugs.clusterlabs.org
> 
> _______________________________________________
> 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://bugs.clusterlabs.org
>