[GE users] Broken queue - pe_hostfile permission denied

Duncan Mortimer duncan at fmrib.ox.ac.uk
Tue Apr 25 16:46:27 BST 2006


Oh well, turns out not to be due to subordination. We've removed all  
of the subordination (changing verylong.q to suspend on CPU load) and  
now the errors are appearing on the long.q  as well as the short queue.
We've tried setting the 'KEEP_ACTIVE' execd_params setting, but  
currently the processing nodes don't seem to be keeping the job  
directories around (the execd's have been soft_stop'ed and restarted).

Duncan
On 24 Apr 2006, at 14:50, Duncan Mortimer wrote:

> Hi,
>
> Further update, we think it may be due to the subordinate queue  
> setting. Removing verylong.q from the list of short.q's  
> subordinates seems to have got things going again, but obviously  
> we'd still like to track down why this is causing this problem. Any  
> ideas?
>
> Thanks,
>
> Duncan
> -- 
> Duncan A B Mortimer DPhil MChem
>                 Computing Officer, FMRIB Centre, University of Oxford,
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
> For additional commands, e-mail: users-help at gridengine.sunsource.net
>

-- 
Duncan A B Mortimer DPhil MChem
                 Computing Officer, FMRIB Centre, University of Oxford,
               John Radcliffe Hospital, Headington, Oxford OX3 9DU, UK.
Tel: (0)1865 222713                             Mobile: (0)7748 105057
WWW: http://www.fmrib.ox.ac.uk/~duncan    email: duncan at fmrib.ox.ac.uk


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list