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

[Condor-users] JobPrio related



Hi,
I have a very basic setup of a cluster and in that I am trying to simulate
the scenario where higher priority jobs preempt lower priority ones (from
the same user). So, assume that I have 20 machines, and I submit 20 jobs
(all having the priority of 0). After this, I submit another job and I
increase the priority of this new job by doing condor_prio (let's say I set
it to -5). My expectation was that this new job would preempt one of the old
jobs and start running. But that didn't happen and it remained idle for
ever. I am sure I am missing something but am not able to find out what.
Could someone please help me here?
In my configuration, I have the following:
WANT_SUSPEND = FALSE
WANT_VACATE = FALSE
START = TRUE
SUSPEND = FALSE
PREEMPT = FALSE
PREEMPTION_REQUIREMENTS = TRUE
RANK = 0

If I do a condor_q -better-analyze, I see "20 match but are serving users
with a better priority in the pool" and 0 for everything else.
Thanks,
Devaraj.