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.

March 2009 log

Primarily CMSSW release installs and removals. Some attempts to debug outstanding issue with CRAB submission via gLite scheduler to UMD.

March 31, 2009

MT -- Removed CMSSW_2_2_1 -> 2_2_4, installed 3_1_0_pre4

MT -- Removed CMSSW_3_1_0_pre2

MK -- Installed CRAB_2_5_0, modified OSG config.ini

Tested OK with condor scheduler, may have same problem with gLite scheduler - never seems to actually submit.
Changed se_control_version to 2.2.0 from 2 (OSG 0.8 style). Stopped services, called configure, and started OSG services. No problems with OSG service starts. Will need to check back with MyOSG GIP to see if we pass tests.

MK -- Installed CMSSW_2_2_6

MK -- Tested SE stageout with CRAB

Jeff attempted submitting jobs to UMD using the condor_g scheduler. Jobs ran but didn't stage data back out to UMD SE. This is a recent problem - I have tested SE output with CRAB on numerous occassions.

MK -- Tested gLite CRAB submission

CRAB gLite submission to UMD is not working from CERN or from UMD. CRAB claims it has submitted the job, but it stays in the Scheduling? state for forever, until eventually it gives up and fails. Watching the cluster, it is apparent that the job is never arriving at UMD. While I have tested gLite submission in the past, I'm not 100% positive that I verified the jobs actually ran at the site - I may have assumed that successful completion via condor_g and successful submission via gLite (according to CRAB) was good enough. gLite submission to UMD has clearly been down for some time - since Jeff first reported the problem over a month ago. Not clear if it's been down since the beginning though. Problem is going to require some serious investigation.

MK -- Removed and installed CMSSW releases