On Mon, Apr 21, 2008 at 7:05 PM, Luis Motta Campos
<[EMAIL PROTECTED]> wrote:
> Xinwei Hu wrote:
> > 1. Change the rsc_colocation to
> > ""
> > <rsc_colocation id="database-disk_together_with_db-cluster-service"
> > from="database-filesystem" to="database-disk" to_role="master"
> > score="INFINITY"/>
> > ""
> > Don't colocate with group.
> > 2. default-resource-stickness = 0
> > 3. database-filesystem's resource-stickness = 1000
>
> Now, I understand what I was doing wrong. A Resource Group *is* *not* a
> resource: hence, it should not be used as a resource in a constraint -
This is incorrect and definitely not your problem.
You're giving the cluster conflicting requirements.
On the one hand, your constraints say the group can only run where
drbd is in master mode.
On the other, you're telling it to "stick" on the backup node where
drbd _ISNT_ in master mode.
Figure out what you want the cluster to do.
> it doesn't make sense.
>
> The problem is that there is nothing in heartbeat that complains of logs
> about this - heartbeat allows me to use a Resource Group name in place
> of a resource in the <constraints/> section of the configuration without
> saying anything. Maybe that's a documentation bug also?
>
> Mentioning the database-filesystem Resource instead of the
> database-cluster-service Resource Group in the <constraints/> section of
> my CIB configuration solved the problem I have, in a consistent and
> understandable way.
>
> Thank you very much for this, Xinwei.
>
> Kindest regards.
> --
>
>
> Luis Motta Campos (a.k.a. Monsieur Champs) is a software engineer,
> Perl fanatic evangelist, and amateur {cook, photographer}
> _______________________________________________
> Linux-HA mailing list
> [email protected]
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
>
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
