PDSF Users Meeting 12/22/09
Attending: Eric and Jay from PDSF and users Andrei, Marjorie, and Jeff P.
Cluster status: Utilization has been relatively light, mostly STAR and some ATLASUCI jobs.
Outages: Discussed slowness caused by atlasusi jobs and the possibility of making /common read-only but no decision on how to do that was reached. The issue of using the NERSC global homes was also discussed as a possibility in the future.
Upcoming downtimes: System-wide outage has been rescheduled for 1/11/10. STAR and ATLAS work scheduled were discussed. The outage will include an SGE upgrade and conversion of interactives to SL5 if time permits. The SGE upgrade is thought to fix the job array accounting bug.
New hardware: Storage is online now and the "diskvaults" webpage has been updated. Order has been place for a dedicated data transfer node.
ATLAS grid node: Need to check w/Iwona on her progress.
Squid/fuse: Jay to send email update later. Squid is working well in Yushu's office for the time being.
- Old qlogin, qsh tickets: qlogin problems seem to be fixed, not clear about qsh.
- STAR project accounts testing: There were problems working on pdsfgrid4 so they would like to get test on pdsf5 instead. Jeff was to file a ticket about this and it will be on the agenda for future meetings.
- STAR running on planck: STAR is interested in trying to run on planck. The easiest way is expected to be via grid submission. Jeff to submit a ticket about the issue.