[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RCF Liason Meeting 123103
(A relatively short meeting today...)
Martin Prushke showed up 00:30 at the counting house to initiate 24x7
coverage. From Jan. 1, 00:30 to Jan, 2, 07:30 Phenix will be receiving
all calls.
PANASAS: a new kernel was installed on a single machine, tests
performed. General consensus is PANASAS is not ready for prime time.
Improper or nonexistent implementation of file locking is apparent cause
of failures - apparently error can be reproduced at will. Recommend
waiting for production version of PANASAS before making final decisions
on deployment.
NFS/ kernel bug: 35 machines went down yesterday, all PHENIX LSF jobs. A
similar set of users were active at time of failures, as was observed
last week - recent attempts to contact these users have so far been
unsuccessful. File moves & deletes apparently not completed when when
NFS automatically unmounts at end of job. New email planned to
phenix-off-l to heighten awareness, and recommend specific job
procedures for submission.
A presentation by Tom Wladek regarding current status of CRS. Problems
with PHENIX executables awaiting scheduled resolution Jan 7-8 timeframe.
Migration to openAFS waits pending (early?) February SLAC meeting and
workshop covering "best practices". Since TransArc version will not be
supported, migration to openAFS is regarded as inevitable.
<<please post to phenix-comp-l or email to sohare@bnl.gov any comments,
corrections, additional detail... thanks & HNY>>