[Linux-HA] DC

[Andrew Beekhof]

On 11/13/06, Zachár Balázs <zachar at direkt-kfki.hu> wrote:
> In my case, maybe there is a reason to change the DC:
>
> In my cluster (active/passive 2 node) the B node is the default DC.
> If I don't use rsc_location constraints the resources will start on the
> B node (I think, It is because the B is the DC).

nope - a node being or not being the DC has no effect on where resources go.

> So I use constraints to
> start they on the A node.
>
> If I use clone pingd. There is a strange thing when I starting the
> heartbeat:
> The resources are starting in the A node, but there is a state when the
> B node's pingd value positive (in my case 110) and the A node's value is
> zero

do you know why this is?

> and then the cluster failover. After the A node's pingd value
> become positive (110) the resources migrating back.
> All of this action were during the start method.
>
> Now I use the ha.cf respawn directive to using pingd.
>
> It is normally?
>
> Thanks,
> Balázs
>
>
>
> Andrew Beekhof írta:
> > On 11/11/06, Zachár Balázs <zachar at direkt-kfki.hu> wrote:
> >> Is there any significance which node is the DC?
> >
> > only internally.
> >
> >> Can I affect this?
> >
> > not really, but then again there's no real reason to need to
> > _______________________________________________
> > Linux-HA mailing list
> > Linux-HA at lists.linux-ha.org
> > http://lists.linux-ha.org/mailman/listinfo/linux-ha
> > See also: http://linux-ha.org/ReportingProblems
> >
> >
> >
>
>
> _______________________________________________
> Linux-HA mailing list
> Linux-HA at lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
>