Re: [HTCondor-devel] Scheduler reports MAX_JOBS_RUNNING reached when it hasn't been

[Ben Cotton <ben.cotton@xxxxxxxxxxxxxxxxxx>]

On Wed, Nov 13, 2013 at 3:35 PM, Todd Tannenbaum <tannenba@xxxxxxxxxxx> wrote:

> Another thought: Could it just be observation skew?  I.e. many shadows are
> exiting at the same time jobs are starting, and/or jobs are switching
> between running and idle very quickly?
>
It's a good thought, but no. There were 1901 jobs in the submission
and it was the only one active. The running job count ramped up to
1899 pretty quickly and stayed there. Job run times average about an
hour.


Thanks,
BC

-- 
Ben Cotton
main: 888.292.5320

Cycle Computing
Leader in Utility HPC Software

http://www.cyclecomputing.com
twitter: @cyclecomputing