[Linux-HA] Groups vs colocations.... etc

[Andrew Beekhof]

On 12/9/06, Andreas Kurz <akurz at sms.at> wrote:
>
> >> Andrew Beekhof wrote:
> >> > On 12/7/06, Andrew Beekhof <beekhof at gmail.com> wrote:
> >> > let me get back to you...
> >> >
> >> > as of this version it should work:
> >> >    http://hg.beekhof.net/lha/crm-stable/rev/1045cec0d37d
> >>
> >> Thanks! I have done some tests with ptest, and it works .... but only
> >> for primitives and not for a group. Is there a special reason for that?
> >
> >i wish i'd never agreed to support groups :)
>
> Why? I really like them :-)

because they're just a hardcoded shortcut and although they work fine
in many cases, as soon people want to do anything a little more
tricky/interesting then groups dont help anymore (and actually make
things worse because its not obvious that they wont work)

> >groups are a non-atomic resource and can't be considered to be running
> >anywhere themselves (groups aren't always collocated although this is
> >the most common configuration).  so any rules that use node attributes
> >will only make sense for the contents of the group and not the group
> >itself.
>
> Yes, that is clear but like groups make all co-location constrainst automatically,
> I thought it would be a convient way to pass my node specific attributes to all
> resources in the group.

and it will for everything except the two stickiness values because
they're calculated differently to allow even this much to work

>
> >
> > Hmmm ... so the score from the location constraint is only added to the
> > first resource in the group.
> >
> >it should pass it on to the rest of them but its effect is not multiplied
> >
> > The default-resource-stickiness is added to
> > every resource in the group.
> >
> >correct
> >
> > The node specific failure stickiness is
> > deleted from the  resources with fail-counts.
> >
> >correct
> >
> > But the result of the
> > group score
> >
> > the group itself doesnt have a score - i'm not sure what you mean here
>
> I mean the sum of the scores of the resources in this group.

that might have been a bug you were encountering.  it seems the
weights of the child resources weren't being merged

>
> > only includes the values from the first resource in the
> > group and resources with a negative score do not initiate a failover of
> > the group like in 2.0.7. So at the moment the failure-stickiness feature
> > works only for the first resource in a group, is that correct?
>
> >if you send me the config i might be able to comment more - it helps
> >to have all the pieces
>
> Oh, sorry .... sounds like a good idea :-)
>
> I've attached my config, I've added the fail counts by hand to that cib
> and did my tests with ptest, version 1045cec0d37d.

if you update again you should find it working a little better

>
> Thanks! Regards,
> Andreas
>
>
>
> _______________________________________________
> 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
>
>