NUC 08 Aug 2012

Marcello Barisonzi; Wolfgang Ehrenfeld; Kai Leffhalm; Shaojun Lu; Hartmut Stadie; Andreas Haupt

id	who	description	status/comments
1007-2	NAF	report on help desk tickets	shown
1010-4	NAF	statistics on multi-core/PROOF usage	shown
1103-1	experiments	test CVMFS, find configuration, communicate setting to NAF admins	CVMFS configured; closed
1204-2	NAF&experiments	rethink queue setups for 15min queue, etc

storage
- lustre still full, moved some users to SONAS, will deplay for all ATLAS users when Yves is back
software
- cvmfs problem (27.6.2012): too large catalog for cache size -> solved by extenting cache partition to now recommended size of 10 GB (for Grid). Not clear why so much catalog information is accessed on WN. Can we diagnose this with the help of ATLAS the next time, if it reoccurs?
operations:
- power outage seen by users, thanks for the user notification and work to get the NAF running again
- no login possible due to full /tmp on prefered WGS (25.6.2012)
NAF tutorial at NAF Germany workshop organized by MB, please inform him for new material, updated procedures

first users for SONAS, problems with jobs in ZN (informed them about SGE option for site selection)
next months CALICE collaboration meeting may cause more users
ILC planning production on NAF
users are happy

id	who	description	status/comments
1208-1	NAF	update storage documentation for SONAS