PASTE meeting

Tuesday 3 Apr 2007, 10:00 → 12:00 Europe/Zurich

Andrei Tsaregorodtsev

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, 3rd April 2007, CERN Chair: A.Tsaregorodtsev Minutes: S. Paterson CERN: G.Castellani, J.Closier, A.Smith, S.Paterson, R.Santinelli Phone: R.Bernet, N.Brook, G.Kuznetsov, U.Marconi, M.A.S.Miguelez, R.Nandakumar, A.Tsaregorodtsev Apologies: M.Bargiotti, P.Charpentier 1. Production Status Report ( J.Closier ) Minutes from the Monday meeting are available here. Joel reported that due to the power interventions at PIC we are currently 'blind'. In the future it was agreed to put a backup mirror of the monitoring pages at another site. Another suggestion was to develop a shell-based monitoring overview which will be pursued with lower priority. In the short term, Marcos offered to put a temporary monitoring webpage in place at USC. ( ACTION: Marcos ) 2. Production Operations Issues - Stager agent and data access status Stuart reported that the Stager is working fine at all sites. Several sites have now successfully run stripping and reconstruction jobs. There was a long discussion regarding the possible causes of the problems at NIKHEF. Subsequent emails have clarified that two problems existed. Firstly gPlazma at SARA was not configured properly to reflect the desired mapping of FQANs to uids/gids (now fixed). The second problem involved an error in the grid map file causing every LHCb user to be mapped to one uid, also now resolved. It was agreed to try to use Joel or Raja's certificate in production over the weekend, requiring a small change to the Agent Director on volhcb02. Tests should be made to confirm this works. ( ACTION: Stuart ) - Data accumulation at CERN and T1 redistribution The plan to accumulate data at CERN and scrap at Tier-1 sites is underway. The list from Marianne was incorrectly compiled for FTS job submission (too many entries for RAL and PIC in the list of replicas). Roberto volunteered to contact Marianne and build the FTS jobs. ( ACTION: Marianne, Roberto ) - Storage and File Catalog clean-up, file removal status ( Marianne absent ) - Investigation of the zero size files registered in LFC Andrew conveyed that his agent can locate and resolve zero size files which have no replicas. There are several cases to consider. If the replicas are of zero size either the transfer could have failed or a zero size file was uploaded. If the transfers succeeded, the LFC registration could have failed. The agent is now running and Andrew will report next week on the progress. ( ACTION: Andrew ) - Proxy handling status Gianluca confirmed that the job receiver now checks the type of proxy with which a job is submitted and will extend this to a default .lhcb VOMS proxy if necessary. It was suggested that the extension should be made based on the original role of the flat proxy. ( ACTION: Gianluca ) The current renewal mechanism will extend a Grid proxy to a Grid proxy, this should be changed to a VOMS proxy. ( ACTION: Gianluca ) Stuart mentioned that although this is expected to be a transparent switch for Ganga users, they should be informed when it occurs. - CS slaves at VO-boxes Raja reported that port 9131 is open on all the VO-boxes and it was discovered during the meeting that the CS slaves are correctly publishing themselves. - User system migration to volhcb01 Andrei reported that the services are deployed on volhcb01 and we can proceed with the migration of the user system. Stuart suggested waiting for Gianluca's fixes before proceeding with the migration such that we have a consistent treatment of VOMS proxies in the WMS, this was agreed. 3. DIRAC3 Tasks - Framework and Databases status Andrei confirmed that the mySQL base class was working. The database definitions have been implemented as discussed at the last PASTE meeting. The changes to the production repository are to be presented in the next two weeks. ( ACTION: Gennady ) - Building tools status Due to the intensive debugging of the reconstruction and stripping jobs, Joel has made no progress with the build tools. There was some discussion over the inclusion of Python in the DIRAC CVS repository with Andrei noting that for SLC3/4 we can pick up the required version from the LCG Applications Area. For more exotic platforms it is not clear how distributions could be built. It was agreed to discuss the issue at the next PASTE meeting with Ricardo. - Setting up lxgate14 for DIRAC3 lxgate14 is now completely free of DIRAC services. The installation of SLC4 and upgrade of the LCG UI etc. for DIRAC3 can now be started. 4. AOB None

10:00 → 10:15 Production status report

- Report from the Production meeting - Monitoring pages backup service

10:15 → 10:45 Production operations issues

- Stager agent and data access status - Data accumulation at CERN and T1 redistribution - Storage and File Catalog clean-up, file removal status - Investigation of the zero size files registered in LFC - Proxy handling status - CS slaves at VO-boxes - User system migration to volhcb01

10:45 → 11:15 DIRAC3 tasks

- Framework status - Building tools status - Databases status - Setting up lxgate14 for DIRAC3

11:15 → 11:25 AOB