NUC 09.02.2011

Hartmut Stadie, Yves Kemp, Wolfgang Ehrenfeld, Johan Blouw, Andreas Haupt, Marcello Barisonzi, Peter van der Reest, Volker Gülzow

- AFS problems:

Understanding of source problem is rather advanced: We think we know the code changes that induced (most of) the problems
need feedback from users
Peter reports on AFS monitoring
will setup a debug server to understand one problematic job
test new job submission verifier
comments from Wolfgang:
- AFS monitoring maybe use batch system errors to monitor AFS

- discussion with Volker:

AFS problems:
- form team to find trigger for problematic ATLAS job
- debug system ready mid of next week
NX/VNC request from ATLAS
- valid use case
- currently not enough manpower to set up and run services
- Volker: will revisit design of NAF, NUC will contribute to this
further questions to Volker: Wolfgang: storage development for ATLAS?
Volker: waiting for realistic data taking scenario to evaluate storage needs

Wolfgang: request maybe 200-300 TB in two months time

- issues of last month

- other issues:

id	who	description	status/comments
1007-1	NAF	setup Lustre groups for ATLAS, CMS, ILC	no progress
1007-2	NAF	report on help desk tickets
1010-2	NUC	resource requests	open
1010-4	NAF	statistics on multicore/PROOF usage
1012-1	NAF	evaluate different stdout/stderr handling for SGE	needs testing
1012-2	NAF	report on CERN strategy for batch and AFS	done, CERN has customized version AFS
1012-4	ILC&NAF	develop strategy to manage Lustre space	ongoing
1101-1	NAF/experiments	WGS reliability/support/monitoring	ATLAS gets more feedback, can understand some problems, access to monitoring could be improved
1101-2	AH/WE/HS	naf_token for ubuntu&MacOSX	not yet
1101-3	NAF	revisit: user information/twitter	no news
1101-4	NAF/CMS/ATLAS	move support list to request tracker?	not yet
1101-6	NAF/experiments	do we need a faster x-connection?	recheck after change of login nodes
1101-8	experiments	test new job submission verifier	will be done
1102-1	experiments	test SL5.6
1102-2	ATLAS/NAF	try CVMFS for ATLAS software	YK: discussed some issues with CERN developers, still not ready to install on NAF nodes; WE: can help with testing it

1104-1	NAF	ease AFS problem reporting
1104-2	NAF/ATLAS	debug AFS problem seen by ATLAS

- Wednesday, May 11th 2011, 1 pm