ATLAS UK Cloud Support

Thursday 29 Oct 2020, 10:00 → 11:00 Europe/London

Vidyo

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

Outstanding tickets

• 149095 UKI-SOUTHGRID-OX-HEP less urgent waiting for reply 2020-10-28 14:02:00 UKI-SOUTHGRID-OX-HEP: unstable transfer
  • Closed prior to meeting; look like issue at the other end.
• 148968 UKI-NORTHGRID-LANCS-HEP less urgent in progress 2020-10-27 15:20:00 UKI-NORTHGRID-LANCS-HEP: deletion and transfer failures
  • Followed up; failing files had already been declared as lost.
• 148342 UKI-SCOTGRID-GLASGOW less urgent in progress 2020-10-27 10:59:00 UKI-SCOTGRID-GLASGOW with transfer efficiency degraded and many failures
  • Awaiting DDM Experts to manually update DB.
• 146651 RAL-LCG2 urgent on hold 2020-10-16 11:56:00 singularity and user NS setup at RAL
  • On hold
• 146374 UKI-NORTHGRID-SHEF-HEP urgent in progress 2020-09-11 13:35:00 ATLAS pilot jobs idle on UKI-NORTHGRID-SHEF-HEP CE
  • on Hold
• 144759 UKI-SCOTGRID-GLASGOW less urgent on hold 2020-08-10 09:54:00 High traffic from UKI-SCOTGRID-GLASGOW on RAL CVMFS Stratum1
  • Decommissioning and physical migration of hardware to new DC.
  • No CEs in old DC.
  • new nat5 to DC soon; then ticket can be progressed.
• 142329 UKI-SOUTHGRID-SUSX top priority on hold 2020-06-04 14:05:00 CentOS7 migration UKI-SOUTHGRID-SUSX
  • Response from Patrick; Host certificate problems

CPU

• RAL

  • Running slightly below fairshare

• Northgrid

  • LANCS: Reduced ATLAS allocation; reduced slots to LHCb

• London

  • Generally fine

• SouthGrid

  • OX system back within 3hrs from Tues downtime

• Scotgrid

  • GLA slowly increasing cores; sofar no bottlenecks seen
  • To follow up with Durham re. reduction in slots.

Other new issues

• Unconfirmed vunerability discussed, to be brought up at UK Security Mtg

Ongoing issues

• CentOS7 - Sussex

  • oh hold

• TPC with http

  • NTR

• Storageless Site tests

  • Oxford agreement for StorageLess tests.
  • O(node(s)) to be allocated for test
  • ~ 2-3 weeks to identify and implement hardware
  • ECDF - remote xrootd monitoring to be udpated
  • For ATLAS:
    • OX - RAL will be endpoint
    • What workloads to start with?
    • Timescale
      • 2-3 weeks
    • Design
      • To be documented
    • Hardware
      • CEs
        • No plan to create new CE; use existing implementation
        • Consider sending a new queue to ARC ?
        • For HTCondor backend; need some mapping
      • XCache
        • Size: estimated 36TB.
    • New Panda Site / Queue likely needed
    • Jira to be created
    • Job Mix.

News round-table

• Vip
  • NTR
• Dan
  • NTR
• Matt
  • NTR
• Alessandra
  • Move to zoom expressed
• Sam
  • NTR
• Gareth
  • NTR
• JW
  • NTR

AOB

• Proposal to Move to Zoom; aim for next meeting.
  • JW to work out how to pass on security information.

10:00 → 10:20 Status

10:20 → 10:40 Ongoing Items

10:40 → 10:50 News round-table

• Vip
  • NTR
• Dan
  • NTR
• Matt
  • NTR
• Alessandra
  • Move to zoom
• Sam
  • NTR
• Gareth
  • NTR
• JW
  • NTR

10:50 → 11:00 AOB

• Proposal to Move to Zoom; aim for next meeting.
• JW to work out how to pass on security information.