[torqueusers] problems receiving output files from jobs

nathaniel.x.woody at gsk.com nathaniel.x.woody at gsk.com
Tue Mar 13 09:32:50 MDT 2007

torqueusers-bounces at supercluster.org wrote on 03/13/2007 11:06:14 AM:

> On Mon, Mar 12, 2007 at 09:05:36PM +0100, Thomas Blum alleged:
> > Hello,
> > 
> > we experience problems with torque-2.1.7(with maui-3.2.6p17) when 
> > submitting 200 jobs in a for loop and getting back the output/error 
> > files for that jobs. The job only does date, hostname and uptime. From 

> > submitted 200 jobs e.g. we get back 198 e and 196 o files. We tried 
> > using scp and rcp but both have the same effects. The missing files 
> > stuck on the clients in the undelivered directories and the serverlog 
> > 
> > Post job file processing error
> > 
> > Does anybody has an idea what we can do?
> The actual error message would have been emailed to the user.  Also, I'd
> check syslog for errors.
> The first thing off the top of my head is "MaxStartups" in sshd_config.

The other one worth checking is ConnectionAttempts in ssh_config. That's 
based on the same assumption as Garricks I believe that you're getting 
something that looks like an ssh connection failure reported back.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.supercluster.org/pipermail/torqueusers/attachments/20070313/6f5020cc/attachment-0001.html

More information about the torqueusers mailing list