NUC 15 Jan 2014

Yves Kemp; Christoph Rosemann; Hartmut Stadie; Alexey Zhelezov

Problems last month:
- on NAF-2 wgs primary group switch not yet consistent, some users report problems on BIRD workers, Proof On Demand does not work with change in primary
- CVMFS: some files were not accessible on some nodes, difficult on detect or monitor
- "-l cvmfs" was not set for all BIRD nodes
- SONAS trouble on 12 Dec, caused by bad disk, contacted support
- ILC wgs heavily swapping on a Saturday, caused by users, fixed with reboot on Monday, experiment admins can contact second level support to get faster reaction time
Grid user interface
- problem: no well working SL6 UI on AFS; removed early Jan 14
- for SL6 install UI locally using PUPPET tool
SL6:
- replace old SL6 setups with PUPPET installations
- have a tutorial on "module" to enable experiments to create their own modules
WN batch migration:
- NAF-1: 2 wgs for ATLAS; 2 wgs for CMS, 27 worker nodes (216 cores)
- proposal: keep this infrastructure until end of March
LHCb supported on NAF-2 and DESY-HH grid
POF-3 review in next weeks, will need user input (papers with major NAF involvement)
network intervention on 11 Feb 5:00 - 8:00 am.

id	who	description	status/comments
1007-2	NAF	report on help desk tickets	shown
1210-2	NAF 2.0	offer better monitoring for queue usage (per user, waiting time, run time, etc)	testing more products, currently investigating other projects; use http://bird.desy.de/status; deploying Ganglia
1210-3	NAF 2.0 & experiments	reconsider queue setup, scheduling, dynamic or configurable queue limits
1302-2	NAF admins	clearly document what scratch space means and policy for Lustre and SONAS	Lustre done

ATLAS is well on the way to moving to NAF-2, I think the majority of DESY users are now migrated.

ILC software will be distributed via CVMFS servers hosted at DESY in the future
Grid UI on SL6 caused problems, will be fixed with new PUPPET installations, where should the certificate be stored?

who

description

status/comments