NUC 13.06.2012

Wolfgang Ehrenfeld; Andreas Gellrich; Andreas Haupt; Yves Kemp; Hartmut Stadie; Johan Blouw; Shaojun Lu

problems:
- CERN AFS problems on June 24th
- CMS CVMFS not working on June 22nd
- slow reading from Zeuthen dCache on June 14th; could not be reproduced
- some hardware issues on WNs
SL 5.8 installed for testing tcx168
additional hardware ordered:
- 16x12 cores for HH
- 180 TB LOCALGROUPSPACE for ATLAS in ZN
SONAS upgrade done last week; still some things to sort out, should be finished end of month

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
1204-2	NAF&experiments	rethink queue setups for 15min queue, etc
1204-3	NAF	short report on current queue setup and usage for June NUC	done
1204-4	NAF	show NAF 2.0 plans in May NUC	done
1205-1	NAF	follow up on slow helpdesk response times	done

Lustre space tight -> waiting for SONAS deployment
ATLAS Support: Period from May 8th to June 13th (Wolfgang Ehrenfeld):
- accounts: ~5
- storage:
  - slow dCache access in ZN, user did not specify where to run (limited CPUs at ZN), moved data to Hamburg
  - AFS
- updated procedure for work group server reboot (inform users on node, inform support, reboot after a working day)
- Lustre full
- CVMFS is now default, no complains
NAF ATLAS discussion mainling list: atlas-germany-naf-users@cern.ch https://groups.cern.ch/group/atlas-germany-naf-users/default.aspx comment from YK: please do not cc mailing lists when writing to helpdesk

replace Lustre with SONAS
some new ideas (for discussion):
- place NAF parts separately inside HH and ZN infrastructure, share common user registry
comment WE:
- please enable to share the AFS home directory between HH and ZN

who

description

status/comments