[Pacemaker] IPaddr2 not failing-over

[Andrew Beekhof]

On Thu, Sep 2, 2010 at 11:44 PM, Ron Kerry <rkerry at sgi.com> wrote:
> So the lrm is obviously issuing the cancel ... but why?

Almost certainly the PE told it to.
The logs should include some lines matching the regex: pengine:.*bz2
If you attach the files referenced by those the lines we'll be able to
figure out what the PE was thinking.

>
> Note that I have also tested this on a full SLES11SP1 cluster running
> SLE11-HAE-SP1 and the ping resources appears to run properly in that
> instance.

Ah!
Then its probably a bug we already fixed :-)
Is Novell still supporting versions of SLE11-HAE prior to SP1?  If so
I guess they have to (find and) backport the fix.

> Therefore, whatever bug is at play here was obviously fixed at
> some point. Right now I am looking for a little more problem definition and
> maybe even the community bug involved so I can go to Novell and get them to
> provide a patch for site that is hitting this problem.
>