Choose timezone
Your profile timezone:
PASTE meeting
→
Europe/Zurich
2-R-014 (CERN)
2-R-014
CERN
Description
CERN, bat 2 R-014
To join the telephone conference, click here: https://audioconf.cern.ch/call/0122929
To dial in to the conference:
a. Dial +41 22 7676000
b. Enter access code 0165202
Production and PASTE Meeting, 1st September 2009, CERN
Chair: S.Paterson
Minutes: S.Paterson
CERN: P.Charpentier, J.Closier, Z.Mathe, S.Paterson, V.Romanovsky, R.Santinelli, F.Stagni
Phone:
Apologies: R.Graciani, E.Lanciotti, M.Rashid, M.Sapunov, A.C.Smith, A.Tsaregorodtsev, A.Zhelezov
0. Ongoing Action List
The only action in the 'ToBeDiscussed' status is regarding the production request page, this should be put into
production such that all PPG users of the page can access it.
SLC5 testing, new binaries should be deployed this week as many sites are migrating already.
1. Production Status Report
- Report from the Production meeting
40M events of the incl_b and incl_c were produced over the weekend, the remaining requests will be defined after the
meeting. The Gauss failure rate is quite high due to the crashes during finalization and this should be investigated.
- Set up stripping workflow(s)
The file naming convention was discussed with Juan and Patrick last Friday and will be tested as soon as something is
available. FSR's are not going to be available on the timescale of the stripping and this should be raised at the Core
software meeting such that some alternative solutions could be implemented. These include: always selecting the
first event or always creating the root output files (event if empty). ( ACTION: Philippe, Stuart )
- Disk quota implementation
Andy sent a mail asking for the storage usage page to be put into production. ( ACTION: Matvey )
3. DIRAC Tasks
- Bookkeeping status
Zoltan commented that the last release did not contain recent modifications and will contact Ricardo to add this to the
v4r19 tag. Philippe mentioned that the BK report module should be revised as several parameters have ambiguous
names and for the stripping the number of input events should be stored. ( ACTION: Zoltan, Joel )
- Release procedure
It was suggested that the release should be discussed weekly, this will become a standing point on the agenda
starting next week.
- Production status report
Updates to the status machine for performing integrity checking of output data are in place but not yet formalized as
agents. As soon as this is done the documentation should be updated.
- VOID card
Stuart mentioned that there was a problem at PIC recently with a few jobs that attempted to download input data.
The amount of scratch space per job was small since 8 cores were sharing the same local disk. On the VO ID card the
scratch space per job was requested as 2GB (insufficient for our merging jobs). This will be increased and it was
agreed to make a review of the VO ID card information. ( ACTION: Joel ) The recent problems with sites imposing a
smaller memory limit on shorter queues was discussed and Roberto agreed to raise the point at the WLCG Operations
meeting. ( ACTION: Roberto )
- VO-boxes
Joel reported that all the VO-boxes were rebooted to cope with the recent security hole found in the operating system.
- Infrastructure monitoring (SAM,SLS,Dashboard)
Roberto commented that storage alarms are now in place and are triggered when 15% is remaining. The SAM suite is
now going to be migrated to Nagios and there is a meeting tomorrow regarding that.
4. AOB
There was some discussion regarding the EGEE demonstration in Barcelona, Roberto will recirculate the abstract.
There are minutes attached to this event.
Show them.