Choose timezone
Your profile timezone:
PASTE meeting
→
Europe/Zurich
Description
CERN, bat 2 R-030
To join the telephone conference, click here: https://audioconf.cern.ch/call/0122882
To dial in to the conference:
a. Dial +41 22 7676000
b. Enter access code 0122882
Production and PASTE Meeting, 28th August 2007, CERN
Chair: A.Tsaregorodtsev
Minutes: S. Paterson
CERN: M.Bargiotti, G.Castellani, P.Charpentier, J.Closier, A.Smith, S.Paterson, R.Santinelli
Phone: R.Bernet, R.Graciani, G.Kuznetszov, R.Nandakumar, M.A.S.Miguelez, A.Tsaregorodtsev
Apologies: N.Brook
1. Production Status Report
Minutes from the production meeting are available here.
Joel reported that there is no major simulation activity occurring. Cedric was asked to clean older jobs from faulty
productions.
Reconstruction is progressing well, almost ~50% of jobs are successfully complete.
Further to the ~800 stalled jobs observed at RAL there was some discussion with Derek Ross about the local dCache
configuration. Apparently the default number of concurrent data connections for dCache servers is very low. This
correlates with what was observed during the reconstruction activity and Derek agreed to increase this limit.
After rescheduling one of the pending CERN jobs Joel observed that the majority have entered the waiting status. A
small number of jobs that have exceeded maximum staging retry are to be investigated. ( ACTION: Joel )
DC04 data cleaning at IN2P3, NIKHEF and GRIDKA is proceeding slowly. Marianne conveyed that the collection of
replica information is nearly finished and the removal of all replicas should be complete by the end of the day. (
ACTION: Marianne ) Once this is over, these sites can be reintroduced to the analysis mask. (ACTION: Stuart )
The LFC RO instance at CNAF was in scheduled downtime recently. This had an adverse affect on services with
repeated timeouts during authentication. In the future this should be removed from the CS in order not to cause the
slowdown. Andrei suggested in the longer term that a 'service mask' could allow more streamlined control of this.
2. Production Operations Issues
Philippe, Joel, Stuart and Andrei discussed a workaround for supporting the SLC4 platform yesterday afternoon. The
standard python can be bypassed and the shared area python may be used instead where available. Stuart reported
that the fixes are now in place such that issues related to 64-bit python at CERN are now resolved. The SLC4 Tier-1
status summary is:
CERN, IN2P3, GRIDKA, PIC - OK
NIKHEF, CNAF - Not available
RAL - Software issues, testing ongoing
For the recent discovery of 64-bit directories in the PYTHONPATH at CERN, Philippe and Roberto agreed to submit a
Savannah bug report ( ACTION: Philippe, Roberto ).
DIRAC v2r15p4 now has to be tested on SLC3 and SLC4 before release. Joel agreed to submit production test jobs to
the available sites. ( ACTION: Joel )
Andrei mentioned that David Bouvet has setup the Oracle LFC instance at IN2P3. There was a question of who to
contact for the configuration of the streaming etc. and LFC support had been contacted. Roberto suggested sending a
mail to the LFC developers list and Andrei agreed to direct the mail there. ( ACTION: Andrei ) Roberto is following up
the problem of the deployment of the LFC distributed service and is in touch with C.Condurache at RAL and D.Bouvet
at IN2P3. ( ACTION: Roberto )
3. CHEP Contributions Discussion
The CHEP contributions page is available here.
It was agreed to discuss the posters in the meeting to facilitate printing. Discussion of the remaining contributions
was postponed until after lunch.
4. AOB
None
There are minutes attached to this event.
Show them.