Hi all,

I'd like to monitor the status of jobs submitted to my LRM via SEG instead of 
polling. In order to acquire some experience and to understand how it works I 
started with fork, but I'm stuck.

I'm using GT 5.2.4 compiled on a SLES 11 machine.
I can submit to jobmanager-fork-seg, as per the default setup in 
$GLOBUS_LOCATION/etc/grid-services.
The scheduler event generator is running under root:

$GLOBUS_LOCATION/sbin/globus-scheduler-event-generator -s fork -d /tmp -p 
$GLOBUS_LOCATION/var/run/lobus-scheduler-event-generator-fork.pid -b

I can see entries in $GLOBUS_LOCATION/var/globus-fork.log and in /tmp/<today's 
date> such as 

001;1368533617;b51461ae-bc8f-11e2-b3dc-0050568f0125:12858;2;0
001;1368533617;b51461ae-bc8f-11e2-b3dc-0050568f0125:12858;8;0

The job runs since I can see the output in the standard output (and if I'm 
correct the last row of globus-fork.log confirms this), but when I query the 
status, for example via globusrun -status, I always get 'active'.

Can someone please point me to I'm missing?

Thanks,

Matteo



Matteo Lanati
Distributed Resources Group
Leibniz-Rechenzentrum (LRZ)
Boltzmannstrasse 1
85748   Garching b. München     (Germany)
Phone: +49 89 35831 8724

Reply via email to