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
    Vidyo

    Vidyo
    Tim Adye (Science and Technology Facilities Council STFC (GB)), James William Walder (Science and Technology Facilities Council STFC (GB))
    Hide

    ● Outstanding tickets

    • GGUS #146538 Glasgow deletion errors. Rod has the important files he wanted off the bad disk. Sam is draining the disk. Is there a tool to say which are the unique files? Sam will discuss with Rod, CC with atlas-support-cloud-uk@cern.ch.
    • GGUS #146525 Sheffield. James will follow up on TB-SUPPORT@JISCMAIL.AC.UK.
    • GGUS #146523 Manchester. Two issues: usual timeouts and file existence check failed. Wait for Alessandra to check.
    • GGUS #146374 Sheffield. No news.
    • GGUS #146280 Lancaster. Matt restarted and it looks better now. Seeing DAVS transfers, but would be surprised if this is production.
    • GGUS #146159 Glasgow. nothing to add
    • GGUS #145688 Manchester. no update from Alessandra.
    • GGUS #145510 RAL now have new WNs with SSD. Seem to be performing much better with IO. Maybe can close this ticket soon.
    • GGUS #144759 Glasgow. no change
    • GGUS #142329 Sussex. Peter applied AGIS updates, but they didn't help. Last email from Patrick to Peter on 13 April. Global setting larger memory fixes jobs, but can't get this set for ATLAS jobs.

    ● CPU

    • RAL new machines installed, giving a bump in running slots.
    • A few planned interventions at some T2s.
    • Gareth: HammerCloud excluded Glasgow due to some missing files. Need to understand.

    ● Glasgow Ceph storage

    • Dimitrios and Sam discussed putting new endpoint online. Will try to bring up this week.
    • Sam thought the AGIS changes looked fine.
    • Naming scheme for the servers: c=cache, s=server; .beowulf.cluster=internal interface, .gla.scotgrid.ac.uk=external interface, but on the same machines.

    ● Grand Unified queues

    No changes. Everything should be done now.

    ● News round-table

    Dan: NTR
    Gareth: NTR
    James; NTR
    Matt: NTR
    Sam: NTR
    Tim: NTR
    Vip: [earlier in the meeting] Discussed making Oxford storageless. Will talk with Pete next week to decide when can go ahead. ATLAS has a well-defined procedure.

    There are minutes attached to this event. Show them.
      • 10:00 → 10:20
        Status 20m
        • Outstanding tickets 10m
          Open ATLAS UK GGUS tickets
          • GGUS #146538 Glasgow deletion errors. Rod has the important files he wanted off the bad disk. Sam is draining the disk. Is there a tool to say which are the unique files? Sam will discuss with Rod, CC with atlas-support-cloud-uk@cern.ch.
          • GGUS #146525 Sheffield. James will follow up on TB-SUPPORT@JISCMAIL.AC.UK.
          • GGUS #146523 Manchester. Two issues: usual timeouts and file existence check failed. Wait for Alessandra to check.
          • GGUS #146374 Sheffield. No news.
          • GGUS #146280 Lancaster. Matt restarted and it looks better now. Seeing DAVS transfers, but would be surprised if this is production.
          • GGUS #146159 Glasgow. nothing to add
          • GGUS #145688 Manchester. no update from Alessandra.
          • GGUS #145510 RAL now have new WNs with SSD. Seem to be performing much better with IO. Maybe can close this ticket soon.
          • GGUS #144759 Glasgow. no change
          • GGUS #142329 Sussex. Peter applied AGIS updates, but they didn't help. Last email from Patrick to Peter on 13 April. Global setting larger memory fixes jobs, but can't get this set for ATLAS jobs.
        • CPU 5m
          UK Cloud jobs over last week
          • RAL new machines installed, giving a bump in running slots.
          • A few planned interventions at some T2s.
          • Gareth: HammerCloud excluded Glasgow due to some missing files. Need to understand.
        • Other new issues 5m

          Oxford to go storageless

          Migrations to Centos7: Lancaster, RHUL

      • 10:20 → 10:40
        Ongoing issues 20m
        • CentOS7 - Sussex 5m
          Centos 7 deployment Twiki
        • Glasgow Ceph storage 5m
          ADCINFR-152: Glasgow Ceph
          • Dimitrios and Sam discussed putting new endpoint online. Will try to bring up this week.
          • Sam thought the AGIS changes looked fine.
          • Naming scheme for the servers: c=cache, s=server; .beowulf.cluster=internal interface, .gla.scotgrid.ac.uk=external interface, but on the same machines.
        • Grand Unified queues 5m
          ADCDPA-235
          Migration plans

          No changes. Everything should be done now.

      • 10:40 → 10:50
        News round-table 10m

        Dan: NTR
        Gareth: NTR
        James; NTR
        Matt: NTR
        Sam: NTR
        Tim: NTR
        Vip: [earlier in the meeting] Discussed making Oxford storageless. Will talk with Pete next week to decide when can go ahead. ATLAS has a well-defined procedure.

      • 10:50 → 11:00
        AOB 10m