Login

ATLAS UK Cloud Support

→ Europe/London
Vidyo

Vidyo
Tim Adye (Science and Technology Facilities Council STFC (GB)), Stewart Martin-Haugh (Science and Technology Facilities Council STFC (GB))
    • 10:00 → 10:10
      Outstanding tickets 10m
      Open ATLAS GGUS tickets

      * 142394    UKI-SCOTGRID-GLASGOW    UKI-SCOTGRID-GLASGOW timeout transfer errors
          Sam: Still some slow transfers from Canada and Japan, but looks like it is recovering. Will check again later today.
      * 142377    UKI-LT2-IC-HEP    High number of hits on ATLAS CERN backup proxy from w*.grid.hep.ph.ic.ac.uk
          Ticket is solved.
      * 142336    UKI-SCOTGRID-DURHAM    Durham to have own squid?
          Adam is looking for hardware. Sam is discussing/encouraging him.
      * 142330    UKI-LT2-RHUL    CentOS7 migration UKI-LT2-RHUL
      * 142329    UKI-SOUTHGRID-SUSX    CentOS7 migration UKI-SOUTHGRID-SUSX
      * 142328    UKI-SCOTGRID-GLASGOW    CentOS7 migration UKI-SCOTGRID-GLAGOW
      * 142327    UKI-NORTHGRID-SHEF-HEP    CentOS7 migration UKI-NORTHGRID-SHEF-HEP
      * 142326    UKI-LT2-QMUL    CentOS7 migration UKI-LT2-QMUL
          will discuss CentOS7 issues later.
      * 142203    RAL-LCG2    RAL-LCG2_MCORE jobs failing
          Two Pilot2/container bugs fixed and reduce failure rate. Still investigating remaining failures.
      * 142136    UKI-SCOTGRID-GLASGOW    High activity from nat005.gla.scotgrid.ac.uk on RAL frontier servers
          Seems to be recovering. See if we close ticket when Gareth returns next week.

    • 10:10 → 10:30
      Ongoing issues 20m
      • Birmingham/Cambridge XCache 5m
        • Cambridge JIRA: ADCINFR-129
        • Sam: XCache working at Birmingham and Cambridge. Mark is on holiday this week. As soon as he is available can set up monitoring with Teng.
        • ATLAS DDM is cleaning up DATADISK and SCRATCHDISK.
      • Diskless sites 5m
        • Sussex JIRA: ADCINFR-130.
        • John Hill: cleanup LGD at Cam and Sussex. Send instructions to John and Patrick (CC UK list).
        • What to do about LOCALGROUPDISK? Tim will send Patrick and John information about what's on their disks. See here: https://twiki.cern.ch/twiki/bin/view/AtlasComputing/UKLocalGroupDisk#Finding_what_data_is_on_your_sit
        • Sussex still looking what to do about SNO+ data. Sam suggested to discuss in the Storage meeting (GRIDPP-STORAGE@jiscmail.ac.uk, http://storage.esc.rl.ac.uk/weekly/)
      • Centos 7 migration 5m
        • https://twiki.cern.ch/twiki/bin/view/AtlasComputing/CentOS7Deployment
        • Alessandra opened tickets for RHUL, Sussex, Glasgow, Sheffield, QMUL.
        • Elena Sheffield CentOS7 migration is mixed up with other migrations (eg. Condor-CE).
        • Elena is looking at AGIS settings for Sheffield. davs seems to be used for some file transfers. Peter proposed to drop davs for everything except deletion.
      • Dumps of ATLAS namespace 5m

        Tim has fixed the dump for RAL Echo. Looking at speeding it up, since it takes nearly 2 days to run, which is problematic as a "snapshot" for comparison with Rucio.

      • xrootd/HTTP transfers 5m

        No news.

      • Pilot2 issues 5m

        No news beyond RAL problems discussed above.

    • 10:30 → 10:50
      News round-table 20m
      • Dan: Making progress on ARC6-CE installation at QMUL. Then will have rest of cluster moved to CentOS7.
      • Elena: NTR
      • John: NTR
      • Matt:
        • Moving to switch off SRM endpoints at Lancaster. Will be discussed at GridPP Technical Meeting tomorrow (https://indico.cern.ch/event/836911/). Matt, Peter, and Tim discussed what we'll say.
        •  Occasional dips in number of jobs at Lancaster. Perhaps due to fixes during the Pilot2 migration.
      • Patrick: NTR
      • Peter: NTR
      • Sam: NTR
      • Vip: Will be away for the next 3 weeks. Pete may be able to look at urgent emails.
      • Tim: NTR
    • 10:50 → 11:00
      AOB 10m
Powered by Indico v3.3.3-pre