Production and PASTE Meeting, 11th December 2007, CERN
Chair: A.Tsaregorodtsev
Minutes: S. Paterson
CERN: M.Bargiotti, G.Castellani, P.Charpentier, S.Paterson, R.Santinelli
Phone: R.Bernet, G.Kuznetszov, U.Marconi, M.A.S.Miguelez, M.Sapunov, A.C.Smith, A.Tsaregorodtsev
Apologies: N.Brook, J.Closier, R.Nandakumar
0. Ongoing Action List
(01) - 06/11/07 - Gianluca, Marianne - Remedy the Exceeded Max Staging Retry jobs and update ProcDB as required
(02) - 06/11/07 - Gianluca, Marianne - Agents for automatic resolution of IntegrityDB entries should be developed
(03) - 30/10/07 - Marianne - Update of LFC replica flags (pending NIKHEF)
(04) - 13/11/07 - Marianne - RAL and PIC migrations, FTS job submissions
(05) - 27/11/07 - Roberto - Oversee LFC intervention plan for upgrade to latest version (testing)
(06) - 27/11/07 - Andrei - Prepare draft pilot document as internal LHCb note
(07) - 27/11/07 - Elisa, Marianne - Prototype bookkeeping with API wrapped in DISET service (volhcb07 will be used for this)
(08) - 04/12/07 - Alexei, Andrei, Gianluca - Further investigate cause of sporadic failures at GridKa for production jobs (pending final tests at GridKa)
(09) - 04/12/07 - Gianluca, Roberto - Prepare an SRM sensor test for the LHCb SAM framework (pending upload to SAM DB)
(10) - 04/12/07 - Gianluca - Report problematic CERN tapes to the Castor support team
(11) - 11/12/07 - Marianne - Clean up SIM/DIGI for production IDs 1957, 1960, 1962
(12) - 11/12/07 - Gianluca - Check User / Production masks and bring up to date
(13) - 11/12/07 - Matvey - Add a flag to lhcb-proxy-init for the length of the proxy
(14) - 11/12/07 - Joel - Update lhcb-proxy-init with small wrapper to ensure environment is not changed
(15) - 11/12/07 - Roberto - Investigate missing myproxy-* commands in 3.1 UI
(16) - 11/12/07 - Joel - Trigger machine upgrades (volhcb06)
(17) - 11/12/07 - Joel - Prepare MDF workflow (pending changes to the job finalization module)
(00) - 11/12/07 - DIRAC developers - DIRAC3 (ongoing)
Completed Actions
Done - 23/10/07 - Marianne, Andrei - Bookkeeping should be installed on volhcb05 and updated to runit instead of init.d startup scripts
Done - 30/10/07 - Marianne - Check number of events for prod 1957 in BK against LFC
Done - 13/11/07 - Roberto - Contact ATLAS representative to understand VOMS usage in other VOs
Done - 27/11/07 - Marianne - Investigate discrepancy of events for productions 1960 and 1962
Done - 04/12/07 - Joel, Andrei, Andrew - Report on upgrade / retirement of older LHCb machines
1. Production Status Report
Minutes from the production meeting are available
here.
The latest stripping was incorrect due to a failure at the application level. Tier-1 status was ok but SARA is down and
rfio instabilities are being observed at CNAF. Gianluca was asked to closely monitor the mask of allowed sites
(particularly for the user system) and bring this up to date. ( ACTION: Gianluca )
Joel reported that the MDF workflow is almost ready but changes had to be made to the JobFinalization module etc. (
ACTION: Joel )
2. Production Operations Issues
- LFC upgrade
The CNAF instance will be upgraded next week but RAL and IN2P3 will not officially be in production before January
2008. NIKHEF, PIC and GridKa were contacted by Roberto and are planning to have LFC instances in place by May.
Roberto will continue testing the new instances when available. ( ACTION: Roberto )
- RAL migration
Stalled, Marianne will submit a small number of FTS jobs to RAL to check the current situation. ( ACTION: Marianne )
- Bookkeeping services on volhcb05
Marianne confirmed that the alias can be changed for the Bookkeeping. The job receiver on volhcb01 can be stopped
and all pending jobs there can be processed in parallel.
- lhcb-proxy-init
A switch should be provided to specify the length of the proxy ( ACTION: Matvey ) and it was agreed to introduce the
small wrapper from Andrei such that the current environment is not interfered with. (ACTION: Joel) Philippe
mentioned that the myproxy-* commands were seemingly not available in the 3.1 UI. Roberto agreed to investigate
( ACTION: Roberto ).
- DIRAC servers usage
Andrei attached the list of currently used servers on the agenda page. After some discussion it was agreed to move
volhcb03 (DIRAC3 development) -> volhcb02 by next week. volhcb06 will be used as CERN-log SE therefore Joel will
trigger the installation and update of this machine. (ACTION: Joel) volhcb07 will be used for BK prototyping. Andrew
requested to keep lxgate34 static since it is running DIRAC2 Data Management services based on a deprecated WMS
distribution. lbnts2 should be scrapped since volhcb07 will become the new BK test instance. volhcb01 will be
upgraded to SLC4 after the migration of CERN-log to volhcb06.
-volhcb01 problem with uberFTP and VOMS proxies reported by Roberto, Maarten Litmaath was triggered to
investigate.
3. DIRAC3
Stuart reported that the full chain is working but there was a problem with recreating the workflows from XML (now
resolved by Gennady). Input data resolution and data upload will be inserted into the Job Wrapper using the new
Replica Manager being developed by Andrew. The API for Gaudi Application jobs will also be written this week.
Joel reported that several Gaudi Application step workflows are running locally and the next step is to introduce linked
input and output parameters. After this, testing can begin.
Matvey has improved the job monitoring and is becoming more proficient with Ajax. Andrei is working to port the
proxy renewal functionality to DIRAC3 and this should be finished this week. Release management tools have been
made available by Joel such that the distribution can be rapidly rebuilt as required.
Philippe queried the progress with using the LCG AA versions of the software. It was agreed to arrange discussion
with Hubert in January. For now a temporary measure will be to include all installed tools in contrib.
4. AOB
None
There are minutes attached to this event.
Show them.