Go to the Indico Home Page Oldest event Older event Up to category Newer event Newest event Printable version
 
Focus on:  Hide Contributions
  • Compact style
  • Indico style
  • Indico style - inline minutes
  • Indico style - numbered
  • Indico style - numbered + minutes
  • Indico Weeks View
    • Switch to the management area of this event
    Login

    EGEE-OSG-WLCG operations meeting

    → Europe/Zurich
    VRVS

    VRVS
    Nick Thackray, Jamie Shiers
    Description
    VRVS: "Forest" room (available 15:30 until 18:00 CET)
    • Monday 10 April
      • 16:00 → 17:30
        28-R-15

        28-R-15
        • 16:00
          Feedback on last meeting's minutes 5m
          Minutes
        • 16:15
          Review of actions items 20m
          actionlist
        • 16:35
          RoundTable 20m

          - ROC/Site reports can be found here: https://cic.in2p3.fr/index.php?id=roc&subid=roc_report&js_status=2
          - VO reports can be found here: https://cic.in2p3.fr/index.php?id=vo&subid=vo_report&js_status=2

          • <big>Asia-Pacific (ASGC)</big> <b><i> </i></b>
            PPS: LCG-CE, WN, MON deployed; glite-CE installed when RC2 released
            CMS SC4: computing model validation, ASGC take ~1000 jobs from Belgium users, since last Thu., and just complete all tasks this afternoon.
          • <big>Central</big> <b><i> </i></b>
            1. R-GMA problems appears from time to time.
            2. There's a problem with site SiGNET failing R-GMA constantly due to software bug. The bug is promised to be fixed in glite 3.0. We've created a ticket in GGUS https://gus.fzk.de/pages/ticket_details.php?ticket=7638
          • <big>CERN</big> <b><i> </i></b>
            Nothing to report.
          • <big>France (IN2P3)</big>
            No report
          • <big> Germany/Switzerland </big>
            - Please notice that messages in ROC report (https://egee.in2p3.fr/CIC/index.php?id=roc&js_status=2&roc_page=5&subid=roc_report&id_report=664) all come from *different* RCs in region DECH. Is it possible to distinguish them by name (maybe adjustment to ROC form)?
            - There's a need in our region for tarballs for mware services other than UI/WN, e.g. SE [main reason: SRM fileservers on different Linux system (debian)]
          • <big>Italy (CNAF) </big>
            No report
          • <big>Northern (NDGF, NIKHEF/SARA)</big>
            No report
          • <big> Russia </big>
            Nothing to report.
          • <big> South East </big>
            * Site TECHION-LCG 2 is having troubles with its condor batch system and LCG 2.7.0. If any site has a similar configuration that has succesfully upgraded to 2.7.0 please let us know.
            * I think its important to collect knowledge about all this efforts in our wiki. Thus I would like to encourage all of you to ask your sites to contibute their know-how to the wiki.
            * SEE ROC has decided to provide at least 50% of its resources to BioMed DC
          • <big> South West (PIC)</big>
            No report
          • <big> UKI (RAL)</big>
            1) One site seeing an issue with VO job use of /tmp on worker node even though EDG_WL_SCRATCH=/data. Is this something we can stop?
            2) Several questions/issues have arisen regarding the OPS VO (agreed at 21st Feb ROC manager's). Many of our sites will deploy it only during the next middleware upgrade.
            3) UKI TPM supporters had first shift last week. Their appear to be several sustainability issues with the current process that we will take to the GGUS TPM group after our shift this week.
          • <big> US-ATLAS (BNL)</big> <b><i> </i></b>
            No report
          • <big> US-CMS (FNAL)</big> <b><i> </i></b>
            No report
          • <big> TRIUMF</big>
            No report
          • <big> KNU</big>
          • <big> Alice</big>
            Alice at this moment has 2 requirements:
            1. Migration of VOBOXES to Itanium machines. They have a large number of sites with this system and they would like to use them. The VOBOX sohuld be therefore compatible with this machines
            2. Begin to run a few number of transfers T0-T1 as alice without interferences with Maarten tests
          • <big> Atlas</big>
            No report
          • <big> CMS</big>
            CMS is in the final three weeks of analysis activities for the Physics TDR. This has resulted in a large number of analysis submissions to FNAL, CNAF, CERN, and DESY. The job abort rate at FNAL over the last 7 days has been abnormally high. There was a configuration problem in many of the user jobs that resulted in job hangs and subsquent job termination and removal, which may have contributed to the abort rate.
            CMS has a sample of a few TB of new Event Data Model (EDM) data. We will begin to replicate this between sites. The sample will be the first large scale test of the trivial file catalog at multiple sites. The data replication step is not anticipated to be disruptive due to the small quantity of data. Accessing this data is not expected to be a large use of resources.
          • <big> LHCb</big>
            No report
          • <big> Biomed</big>
            No report
            => 2nd BioMed Data Challenge (https://edms.cern.ch/document/711157)
          • <big> GGUS TPM</big>
        • 16:55
          Status of Pre-Production Service 5m
          Speaker: Nick Thackray
        • 17:00
          Review of Experiment Plans and Site Setup 5m
          Speaker: Harry Renshall
        • 17:05
          Status of dTeam Transfers / SC4 Throughput Tests 5m
          See SC4 disk-disk transfer status for more info.
        • 17:10
          AOB 5m
        • Next meeting will be on Tuesday 18th April at 16:00 CET