How To: Guides for users and Maryland T3 admins.

Help: Links and emails for further info.

Configuration: technical layout of the cluster, primarily for admins.

Log: Has been moved to a Google page, accessible only to admins.

July 2009 log

Fixed condor configuration so that jobs can be suspended but not evicted for performance or priority reasons. Installed SuperB FastSim software temporarily, waiting on decision from Nick whether we will allow them permanent installation access.

To Do & Sandbox

July 23, 2009

MK -- Installed SuperB FastSim software on phedex node

To temporarily allow SuperB folks to compile their code, installed software according to these instructions.
Will revisit this permissions issue with Nick.

July 21, 2009

MK -- Garbage collection on WNs

Added weekly garbage collection of directories on WNs inside /tmp as well as edited kickstart file for the same.

July 15, 2009

MK -- Tweaked condor configuration

Condor jobs still getting evicted (suspension is OK - job stays in memory, so CMSSW jobs don't have to restart from 0 every time they're suspended), so tweaked condor configuration to:
PREEMPTION_REQUIREMENTS = False
NEGOTIATOR_CONSIDER_PREEMPTION = False
CLAIM_WORKLIFE = 300
WANT_SUSPEND = True
SUSPEND = ( (CpuBusyTime > 2 * $(MINUTE)) \
&& $(ActivationTimer) > 300 )
CONTINUE = $(CPUIdle) && ($(ActivityTimer) > 10)
PREEMPT = False
Initially set CLAIM_WORKLIFE to 1, but condor did weird stuff - number of running jobs and claimed slots fluctuated wildly. Suspect that jobs weren't actually starting and schedd was relinquishing claim before the job ran. Set to 300 and now jobs doing much more sensical stuff. schedd does seem to be relinquishing claim as load increases, but not at ludicrously low numbers. Restarted condor services a couple times and ran tests - latest jobs behaved well.

July 2, 2009

MT -- Removed CMSSW_3_1_0_pre7, installed CMSSW_3_1_0

July 1, 2009