Choose timezone
Your profile timezone:
Regional Operations Meeting DECH
→
Europe/Zurich
FZK
FZK
Description
Telephone conference. Details for connecting to the conference were distributed through the regional mailing list.
Minutes for Regional Operations Meeting DECH (January 26th, 2007)
Attendance:
Clemens Koerdt <chair> (FZK)
Alessandro Usai (CSCS)
Andreas Gellrich, Uwe Ensslin (DESY-HH)
Kläre Cassirer, Ute Karabek, Jisamma Kallumadikal, Horst Schwichtenberg (SCAI)
Hans-Gunter Borrmann (Uni Freiburg)
Apologies:
(Uni Wuppertal)
(RWTH-Aachen)
(ITWM)
(GSI)
Missing:
(MPPMU)
(DESY-ZN)
(Uni Karlsruhe)
(Uni Dortmund)
1. Introduction
- Announcements:
o Update 12 released, priority: normal
o A critical bug on the gLite-WMS version contained in update 12, see: http://glite.web.cern.ch/glite/packages/R3.0/updates.asp
o Reminder: APEL2 available since update 10; Support added for glite CE and SGE batch system. Fixes of various bugs.
o PPS: Update 16 released.
o GridView service is now available also for PPS sites. The site view is a good tool for site administrators to verify the overall status of a site (http://gridview.cern.ch/GRIDVIEW/same_index.php)
o Sam admin page integrated into CIC portal: http://cic.gridops.org.
o Reminder: Operations web site: www.gridops.org (Gstat, cic-portal, gocdb, GOC-home, sam, sam pps, sam admin page, gridice)
o EGEE All Activity Meeting (see links on agenda page)
Issues: WMS, gCE not yet production quality, Migration to SL4/64bit progressing slowly, different version of CE – need of harmonization.
o Status EGEE III proposal (see link on agenda page)
- Action Items (partner review further below)
o PPS: GSI added resources (#! ITWM not yet), GSI to get infosystem working and SAM tests for one week before being certified. -> GSI, ITWM not present at meeting
o Dech VO: Supported? What sites do support? -> Voms server at SCAI is not working, needs perhaps to be reinstalled. Some more sites were working on the issue of supporting the Dech VO, but could not be tested.
o SFT-Server: -> no further progress, need to address filtering of results.
2. Round the Sites
CSCS
----
o updated m/w (U12)
o experienced some problems with maui configuration file
o trying to have new cluster in production us soon as possible
o dcache might take a bit longer (end of February?)
o disappointed about remaining problems with SL4 WN packages
o now trying to make cluster work under SL3
DESY
----
o Update 12 done
o bug fix was also put in
o running one WMS in parallel to 3 RBs
o found still some issues with new (‘intermediate official’) SRM version (as part of dCache developed at FNAL).
o incidence with ‘screwed’ certificate: resulting in damaged gridmap files and was a medium disaster; turned out to be a bug in gLite (handling of certain characters in the DN string..).
o PPS has been running smoothly.
DESY-ZN
-------
...
FZK
----
o still struggling with rm-test failures after load increased.
o using dCache workshop at DESY and WLCG collaboration workshop to sort things out
GSI
----
….
ITWM
----
(report received by mail after the meeting):
status production:
- system running, load quite high.
- failing Sam tests for some time. reason was system load in combination with MAUI scheduler, failing to start the Sam jobs on time. reconfiguration of Maui solved issue.
status PPS:
- softwareinstallation on existing nodes (z.zt. 10 maschinen)
- at the moment integrating additional machines into the EGEE cluster, for both Production and PPS usage.
SCAI
----
o Installed updates
o Ute Kabarek will leave SCAI. (All of us wish you good luck and fulfillment in your new job!)
o minor problems with load gatekeeper processes appearing, now fixed.
o user problem: want to run biomed very long jobs >24h. But after 24 hours jobs are getting aborted. Longer proxies were requested, but it seems as if a configuration specific to the Biomed VO is responsible. Problem was forwarded for discussion to the rollout list.
CK suggested to also opening a DECH ticket for our internal mw-deployment support unit in order to address upcoming problems within region (first).
Dortmund
--------
….
Karlsruhe
------
….
Freiburg
----
...
3. Feedback to the TCG
--------------------------------
o no particular issue presented at the meeting
o GSI and SCAI planning to give a call to every resource center within the region to collect issues
4. COD
--------
o New duty shift coming week.
o Questions will be how to share workload with backup team?
o COD-12 will take place in Karlsruhe 6-8.2.
5. ROC-On-Duty
------
Handover CSCS to DESY:
o CSCS: not many tickets, interesting progress on ticket concerning maintenance of RB (#1237), procedure proposed how to dump parts of ever growing database.
o DESY: procedure not entirely satisfying, only partially successful when trying their suggestions…. (One gets the feeling that you are always the first who uses components in production….). We have given up on issue how to drain RB. Things do not seem to have improved with the WMS system, which takes a lot of effort administrating…,
o DESY: Overall backlog of tickets has decreased, which is nice.
5. Discussion Partner Review
----------
o Reminder: THIS IS A PARTNER REVIEW, not a roc review! (So it's in your own interest to present yourself in the best possible way!).
o Received material from all partners, but in parts unfinished/preliminary/short ?!?
Please review thoroughly your contribution and extend/change if necessary!!
o Please upload Logo and photo from your resource center to Wiki page, if not done already. (!# CSCS, ITWM, SCAI )
Questions, remarks:
HS: Please give us feedback if anything is missing.
CK: This is continuously done so, if anything specific is found. (But substance can not be added by partner).
HS: How can we compare our contribution with that of the others?
CK: Draft will be put on DECH-Wiki soon. (but form is not problematic)
o HS: What about the funded/unfunded issue
!#SH to clarify. (This was done at the time of writing: Sven has raised this issue with Alistair Mills. Short term outcome is that this (financial) point is not examined at the coming partner review. More details will be given in the medium term (after consultation with Marcel Kunze).)
o HS: we are preparing a slide about the Dech VO and will send a version around for comments soon (!#HS). Any resource center within our region that is using the Dech VO for applications, please come forward to include this in the slide. (!#all)
o It was suggested to synchronize ourselves on (how to present) a common DGRID strategy for the review.
o CK: we at GridKa aim to use D-Grid and EGEE resources in common. Could that be a common goal within our region?
o HS: At SCAI we currently have two different clusters. And bringing them together should be seriously discussed first….
o AG: Not a priority for DESY. We don't know the interference; LHC is coming soon with big milestones involved, and unstable services. Not a very good time to play around… In addition, things in D-Grid do not seem very much advanced.
o CK: It was good to have started the discussion, but for the moment there is no need to go further at this point (since we will probably not reach a common vision in time for the review). Anyhow, it will surely not be the last time we discuss the issue here…
6. AOB
------
Reminder to register for meetings in Karlsruhe in case you intend to participate…
There are minutes attached to this event.
Show them.