Go to the Indico Home Page Oldest event Older event Up to category Newer event Newest event Printable version
  • 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

    ATLAS UK Cloud Support

    → Europe/London
    Zoom

    Zoom
    James William Walder (Science and Technology Facilities Council STFC (GB)), Jyoti Prakash Biswal (Rutherford Appleton Laboratory)
    Description

    https://cern.zoom.us/j/98434450232

    Password protected (same as (new) OPs Mtg)

    Hide
    There are minutes attached to this event. Show them.
      • 10:00 → 10:20
        Status 20m
      • 10:20 → 10:40
        Ongoing Items 20m
        • New Issue(s) 20m

          • Data Challenge test transfers have started (Tier-0 export, configured with 20 Mbps each T0->T1 channel).

            • 17 Mbps were observed in the monitoring.
            • Link1: https://monit-grafana.cern.ch/goto/kGmfKZ7SR?orgId=17
            • Link2: https://gitlab.cern.ch/atlas-adc-ddm/dc_inject

          • LANCS looks happy so far (after the migration)!

          • AMI/Rucio synchronisation - datasets are empty in Rucio and fuill in AMI.

            • Fixed by AMI team.

          • Gateways have been upgraded at RAL: svc11, 13, 14, 15.

            • In Cluster Management Service Cluster (CMSD).

          • ECDF_CLOUD update.

            • After comparing noted with the CS team, there aren't any problems from the CS side, no stuck VM, no failing brokering, no difference in config between ECDF_CLOUD and other sites.
            • Do we know if this is XRootD or davs transfers which are failing?
            • Could the _CLOUD queue be attempting to use a different protocol for file access/transfers against GLASGOW? If not, try changing it to use a different protocol and see if there are less issues?

          • DURHAM joblessness.

            • Looks like the Pilots just don't have any work to pull in, it could be ATLAS not having enough work at the moment.
            • LHCb looks good though!

          • RAL-LCG2_MCORE brokeroff is set to retire.

            • Prod/MCORE PQ used to help scheduling, which was unhappy at following gshare when switching S-MCORE and back. Assume scheduling better now.
            • Partitioned resources bad for latency and following gshare.

          • CPU efficiency @BHAM (from last week).

            • A couple of broken disks were replaced on two of the worker nodes and they both would have done a RAID rebuild.
            • Things seem to be OK again now!
          Commissioning new SLURM/EL8 ARC-CE at UKI-NORTHGRID-LANCS-HEP
          GLASGOW-monit
          Link1
          Link2
        • Storageless Site tests/Decommissioning 20m

          MAN

        • Localgroupdisk cleanup and migrations 20m
        • Regular file dumps 20m
          Twiki
      • 10:40 → 10:50
        News round-table 10m
      • 10:50 → 11:00
        AOB 10m
        Mattermost
        TWiki:StorageSetUp