Core Minutes 26 May 2015

Core Minutes 5/26/2015

Present: Joanne Bogart, Jim Chiang, Richard Dubois, Warren Focke, Heather Kelly, David Landriu, Leon Rochester,

ScienceTools: (Jim) No news.

See Science Tools Development Notes for current activity.
Hardware: (Richard for Tom) The switchover to the new gpfs user disk is complete. There were some issues:
- Stale nfs mount points. This was expected. They were all cleared up within a day.
- A conflict between native GPFS clients on certain batch nodes, and the cNFS layer which prevent users from seeing their spaces. Fixed by disabling the Fermi "fs2" file system to the GPFS clients.
- An issue with file locking. GPFS/cNFS support NFS v3 and not (yet) v4 which may be an issue for sqlite3 (used by iPython). Being investigated.
(Richard) In other news, we've ordered 1/2 petabyte of tape.
Pass 8: (Leon) Nothing to report. (Jim) Testing at FSSC is beginning; may be done in a couple weeks. (Richard) Is L1 ready? (Warren) Yes, L1 and dqm are ready for Pass 8.
Core talk:
- Wired and retiring glastlnx07 (Heather) As a step towards running GR with Wired on fermilnx-v14, Heather tried running Wired standalone. She cannot get it to run on her own Windows box, her SL6 Linux laptop, or on rhel6-4 or rhel5-32. The problem was reported to Tony and Dima. They have reproduced the problem and note things were working fine 2 months ago. Richard recalled that part of the issue may be associated with security policies at SLAC - Heather doesn't remember that part from the email exchange. She will contact Tony and Dima again to see if they have further insight.

J. Bogart, Last Modified: