[Condor-users] long running job advise

[Rita <rmorgan466@xxxxxxxxx>]

hi

we currently run jobs longer than 6 days. There have been times when the job dies (shadow_excetion) mostly because I suspect I am doing a stream_output.

I understand the job restarts occur for whatever reason, but are there any ways to just deal with it without loosing too much work? I was wondering how good the check pointing in condor is -- most of my jobs are perl jobs -- or do people recommend setting up a VM instance?

--
--- Get your facts first, then you can distort them as you please.--