[Linux-HA] Question regarding the stonith resources

[Andrew Beekhof]

On 10/31/05, Peter Kruse <pk at q-leap.com> wrote:
> Andrew Beekhof wrote:
> > On 10/31/05, Peter Kruse <pk at q-leap.com> wrote:
> >>
> >>hm, that is a one-node-cluster?  are you saying that
> >>in a two-node cluster it's always one node fencing the other?
> >
> >
> > no - i'm saying that all fencing directives come from the DC.
> > who actually caries out the directive is dependent on how you
> > configured the stonithd (via the placement and configuration of
> > fencing resources)
>
> yes, I can create two stonith-resources and create
> the constraints to make one stonith resource run
> on node A and the other on node B.  I cannot however
> tell the stonith resources that it can only fence
> the other node.  There is no parameter for that.
> So you have to go through try
> and error, I guess.  That is you try to tell node A
> to shoot itself, find out that it can't do that,
> and switch over to node B and tell it to shoot node A, right?

really?
surely there is a way to say (for each stonith resource) "i can shoot
node X, Y and Z" no?

> >>Are you ignoring the result?
> >>
> >
> >               } else if (ST_OK == stonithd_node_fence( st_op )) {
> >
> > and
> >
> >       switch(op->op_result) {
> >               case STONITH_SUCCEEDED:
> >
> > So the CRM is definitely checking the result.  However when the DC
> > shoots itself, it is by definition not there to check the result
> > (unless of course the op fails).
>
> right, can it happen that the DC shoots itself but doesn't succeed
> and the other node still takes over?

no - if its still around - then its still the DC