[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Condor-users] condor_q



Hi all,

I have a pool of 3 machines working fine already. But I still have concerns:

1. How can I see the TOTAL JOBS (with their status) on the pool? Because
condor_q displays the jobs ONLY on the machine where I command condor_q. I
want to know all the jobs running on the pool so that I may know on how
long is my job in the queue line.

2. Another related thing is that, when I tried to submit a batch of jobs
from one machine (say machine 1), they were executed RIGHT AWAY. If I have
another set of jobs sent from the SAME machine, they were executed RIGHT
AFTER the first set of jobs are done. But the problem arises when I tried
to submit jobs from other machines (say machine 2). They just stay IDLE
FOREVER even when the jobs on the first machine were already done. They
can be run only IF and ONLY IF another set of jobs is submitted from
either the same machine (machine 2) or from other machine (machine 1). If
the last set of jobs is submitted from the same machine (machine 2), then
this set of jobs can be executed and everything will goes okay, BUT if the
last set of jobs is submitted from other machine (machine 1), then the
same problem occurs...and this continues...unless of course (my
workaround) the last 2 sets of my job is to be submitted with the last job
as a dummy job just to make sure that the second to the last job is to be
executed.

Thanks and cheers,
Leo