Re: RMFailure

On Wed, May 10, 2006 at 03:20:31PM +0200, Enrico Morelli alleged:
> Dear all,
> 
> I have a cluster of 12 nodes using torque 1.2.0 and maui 3.2.6p13. On
> a node I have a problem, the pbs_mom uses 99% of CPU. Ok, I restart
> pbs on the node and from the server I checknode:

Anything in MOM's log or syslog?  Whatever the problem, it is likely
already fixed in modern TORQUE.

> job is deferred.  Reason:  RMFailure  (cannot start job - RM failure,
> rc: 15041, msg: 'Execution

This means the job couldn't be sent to pbs_mom.


> If I stop pbs_mom under the node that had problem, all jobs runs properly.
> 
> Where is the problem?

You can trace or attach gdb to the running pbs_mom process and get a
backtrace.

But I'd just start off with upgrading TORQUE.