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, 13 February 2007, CERN
Chair: A.Tsaregorodtsev
Minutes: S. Paterson
CERN: G.Castellani, J.Closier, A.Maier, J.Moscicki, A.Smith, S.Paterson, R.Santinelli
Phone: A.Carbone, C.Cioffi, N.Brook, R.Nandakumar, A.Tsaregorodtsev
1. Production Status Report ( J.Closier )
Minutes from the Monday meeting are available here. Joel reported that two test productions for DIRAC v2r13 had
run smoothly. Jobs which appeared to have failed had either not produced the correct number of events due to low
statistics or tried to execute while the recent software installation was still in progress. DIRAC v2r13 can now be
installed in the release area and deployed to the Production system. ( ACTION: Joel ) The VO-boxes should also be
updated to the latest version. ( ACTION: Raja )
Andrei reported that DIRAC v2r13p1 has also been built as a development version for the new job wrapper which
utilises lcg-getturls. This should be extensively tested before release. ( ACTION: Stuart )
2. Production Operations Issues
The VO-boxes have accumulated ~5K requests over the past two weeks and should now be cleaned. ( ACTION: Raja
) Raja reported a bug which results in the Transfer Agent hanging. This should be investigated and understood. (
ACTION: Ricardo )
Andrei reported that the processing DB is the last component of the production system running on lxgate14. This
should be migrated to the new host volhcb02. ( ACTION: Andrei ) All services on lxb2008 have now been stopped.
Andrew Maier has confirmed there are no problems for Ganga with the latest version of the DIRAC API. The User
system can now be upgraded to the latest version and migrated to volhcb01. ( ACTION: Andrei ) It was agreed that
the DIRAC API will continue to be supported on python2.3 after the standardization of the rest of DIRAC to python2.4.
Andrew mentioned that the list of SIM files which can be removed has been circulated and requested data will be
retained. To allow 'bulk removal' it was suggested to ask site admins to remove directories on physical storage and
clean up LFC entries in parallell. ( ACTION: Andrew )
There are still many files in the LFC for which the SE name is incorrect. Marianne should urgently contact the LFC
developers to organise the renaming of these files. ( ACTION: Marianne )
It was decided that the DIRAC construction of TURLs should now be stopped in the Job Wrapper. ( ACTION: Stuart )
The migration of the bookkeeping from lbnts2 to volhcb01 is now a priority in order to retire the machine. Carmine
reported that the migration should go smoothly since volhcb01 is running SLC3 with python2.2. ( ACTION: Carmine )
3. DIRAC3 Tasks
Ricardo stated that the DIRAC3 frameworks were under development. The need for the DIRAC3 CVS repository was
mentioned and Joel confirmed this would be in place by the end of the week. The CVS structure as discussed in
Barcelona is to be circulated. ( ACTION: Joel )
Andrew summarized the current status of the data moving tools for the pit in the following slides.
4. gLexec tests on the PVS
Stuart reported that the Preview Testbed is not currently ready to begin gLite tests. Roberto is to keep track of
progress and confirm when it is possible to start. ( ACTION: Roberto )
5. DIRAC Sites (Syracuse and ICHEC/Ireland)
Andrei mentioned that both Syracuse and ICHEC are non-standard DIRAC sites. There is no batch system at
Syracuse and at ICHEC there are many local settings which complicate the situation. It was suggested to arrange a
phone meeting with ICHEC site in order to resolve outstanding issues.
6. AOB
Joel confirmed he would attend the Tuesday meeting in order to announce the clean up of SIM files.
There are minutes attached to this event.
Show them.