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, 5th February 2008, CERN
Chair: A.Tsaregorodtsev
Minutes: S. Paterson
CERN: G.Castellani, J.Closier, Z.Mathe, S.Paterson
Phone: A.Casajus, G.Kuznetszov, M.A.S.Miguelez, M.Sapunov, A.C.Smith, A.Tsaregorodtsev
Apologies: M.Bargiotti, N.Brook, P.Charpentier, R.Santinelli
0. Ongoing Action List
(01) - 06/11/07 - Gianluca - Remedy the Exceeded Max Staging Retry jobs and update ProcDB via development of Agents to resolve this (merged with Action(02))
(03) - 30/10/07 - Marianne - Update of LFC replica flags (pending NIKHEF)
(04) - 13/11/07 - Marianne - RAL and PIC migrations, FTS job submissions
(07) - 27/11/07 - Elisa, Marianne - Prototype bookkeeping with API wrapped in DISET service after ensuring AMGA and Oracle availability on volhcb07
(09) - 04/12/07 - Roberto, Marcos, Gianluca - Develop LHCb SAM framework and summarize architecture for SE tests
(25) - 08/01/08 - Marianne - Resolve why number of processed events is missing in Bookkeeping (also cause of missing files)
(27) - 08/01/08 - Andrew - Backup DMS services on lxgate34 to allow migration to SLC4
(30) - 15/01/08 - Gianluca - Test NIKHEF after reinstallation of the shared area
(32) - 22/01/08 - Gianluca, Marianne - Identify and cleanup faulty productions with duplicated events
(33) - 05/02/08 - Joel - Check new ProcessingDB functionality and report any problems
(34) - 05/02/08 - Joel - Trigger set up of disk partitions on volhcb01
(00) - 05/02/08 - DIRAC developers - DIRAC3 (ongoing)
Completed Actions
Done - (29) - 15/01/08 - Gianluca - Remove lost files at PIC from ProcDB
Done - (31) - 22/01/08 - Roberto - Investigate cause of CERN RB NS behaviour
1. Production Status Report
Minutes from the production meeting are available here.
- Remedy Exceeded Max Staging Retry jobs and update ProcDB - Action(01)
Andrei conveyed his comments regarding the draft of Gianluca's Agent. There are three different status fields in the
ProcessingDB, firstly for the LFNs, secondly for each physical replica and thirdly for each file in a transformation. The
last two are the important fields and should be set correctly by the Agent that corrects problematic replicas. The
suggestion is to retain the original LFN status and mark problematic replicas correctly. Any files that are not
processed should be marked as unused. To handle the existing problematic replicas in the ProcessingDB it was
agreed to write an Agent to convey this information to the IntegrityDB where another Agent can perform any
corrective action.
Andrei reported that some additional functions have been added to the DIRAC2 ProcDB to improve functionality and
Joel was asked to check these and report any problems. ( ACTION: Joel )
Gianluca added a SAM test to check that the LCG Utils version. If this is older than 1.6.7 the site will appear in
'Warning'.
2. Production Operations Issues
- Backup DMS on lxgate34 - Action(27)
Andrew agreed to pursue this when some free time was available but with low priority compared to CCRC08 tasks.
- PIC lost files - Action(29)
Gianluca mentioned there were no lost files for the ProcessingDB so the action is closed.
- Test NIKHEF shared area - Action(30)
The missing LHCb package library was not yet checked, Joel agreed to convey the software version to Gianluca for
test jobs to be directed to NIKHEF.
- RB NS behaviour - Action(31)
Yvan installed a monitoring tool on the RBs to track the NS on their side. Since last Thursday there have not
been any failures so this action will be closed.
- Correct duplicated events with faulty productions - Action(32)
Gianluca has gathered a list of all jobs with duplicated events using Philippe's script. The list of faulty productions
that must be rerun has been conveyed to Marianne and Joel confirmed these have now been rerun. The Bookkeeping
must be updated in both cases.
3. CCRC Operations Issues
- volhcb02, volhbc01 status
The disk partitions on volhcb01 need to be setup correctly for DIRAC as the bulk of available disk is not available. (
ACTION: Joel ) The installation of volhcb02 can now proceed after the recent upgrade, this can house a mirror of the
DIRAC3 CS and a VO-box.
- DIRAC3 release (and tools) status
The two options here are to prepare a fully tagged release and install via tarball or to install on volhcb01 from CVS
during the initial testing phase. It was agreed to start with the latter to decrease turnaround time for fixes as and when necessary.
- RAW data Bookkeeping
The last report from Elisa is that the Bookkeeping information prepared for RAW data was not properly processed.
- Pit-T0 test status
Andrew commented that all tools for the Pit -> Castor transfers are in place, testing is ongoing.
- T0-T1 transfers
TransformationDB code was checked and Andrew confirmed that only integration and testing work was now required
for this activity to start.
4. DIRAC3
- Data Management status
Andrew mentioned that T1-T1 transfers with the failover mechanism are also pending. T0-T1 transfers use the same
mechanism. The generic File Catalog client is now in place and there have also been fixes to the Replica Manager.
- Workflow status
Joel confirmed that with Gennady last week a typical workflow of multiple steps and modules was successfully tested.
The Gaudi Application and Check Log File modules for Brunel are in place. Andrei conveyed that the Bookkeeping
module is now on the critical path. The Job Finalization module should also be pursued with Andrei and Stuart.
- Production Management status
Gennady commented that the CLI is in progress for basic production management operations.
- WMS status
Stuart reported that the gLite Pilot Director is now working after some minor debugging. The input data policy in the
Job Wrapper has been made generic for all VOs with simple plugin modules. A DIRAC Admin API is now in place and
will start to become more powerful over time. Andrei mentioned that pilot management is also operational, retrieval
of pilot outputs for debugging and presenting this on the webpage is ongoing.
- Accounting and Monitoring status
Adri and Ricardo have been making progress with activity monitoring and accounting. The Job Wrapper and
Workflows should be populated to report to the Accounting service.
5. AOB
None
There are minutes attached to this event.
Show them.