[Pacemaker] did I find a bug in the OCF script for clvmd?

[Glauber Cabral]

Hi Sander.
I'm new to to pacemaker, too.
The OCF_ROOT variable needs to be set before running the script by hand.
When the script is run by pacemaker, the variable is already set (I
don't know where, exactly).
So, the fact that the variable is not set is not a bug.
I guess you should report the problems you get after setting the
variable (if running the script by hand) or the errors from the log
(if running from the cluster stack).

I hope I could help you a little =)

Cheers,
Glauber Cabral

On Fri, Feb 12, 2010 at 7:07 PM, Sander van Vugt <mail at sandervanvugt.nl> wrote:
> Hi,
>
> I'm still struggling with clvmd and trying to understand what's
> happening, see the description I've sent earlier this week. Now I've
> found something more specific, please tell me if I've found a bug or
> not: (using SLES 11 with HAE)
>
> When trying to run the /usr/lib/ocf/resource.d/lvm2/clvmd script
> manually with the command ./clvmd start (from the lvm2 directory of
> course), I ran against an error:
>
> ./clvmd line 27: /resource.d/heartbeat/.ocf-shellfuncs: No such file or
> directory
>
> This error seems to come from a reference to the OCF_ROOT variable in
> the ocf script itself, which seems to be unknown. After adding the line
>
> OCF_ROOT=/usr/lib/ocf
>
> to the clvmd script, at least the error from the default OCF script goes
> away.
>
> I would appreciate if someone more knowledgeable than I am could have a
> look at the script and tell me if this indeed is a bug, or if it's
> getting too late, I've been working too hard and need a couple of beers.
> Lots of beers :-)
>
> Thanks,
> Sander
>
>
>
>
>
> _______________________________________________
> Pacemaker mailing list
> Pacemaker at oss.clusterlabs.org
> http://oss.clusterlabs.org/mailman/listinfo/pacemaker
>