ATLAS UK Cloud Support

Thursday 16 Sept 2021, 10:00 → 11:00 Europe/London

Zoom

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

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

Password protected (same as (new) OPs Mtg)

● Outstanding tickets

• • 154023 USER atlas UKI-SCOTGRID-ECDF less urgent NGI_UK in progress 2021-09-13 16:30:00 High traffic from UKI-SCOTGRID-ECDF to DESY CVMFS stratum 1 WLCG

    • In progress

  • 153550 TEAM atlas UKI-SOUTHGRID-RALPP less urgent NGI_UK on hold 2021-09-01 15:56:00 Transfer failure at UKI-SOUTHGRID-RALPP with “Failed to select pool: All pools are full\n” error WLCG

    • On hold to collect some stats;
    • Should now be progressed

  • 153367 TEAM atlas RAL-LCG2 urgent NGI_UK in progress 2021-08-04 11:55:00 HTTPS on RAL CTA WLCG

    • Needs response

---

● CPU

• • RAL

    • Some variation due to atlas scheduling / capacity

  • Northgrid

    • LANCS; removed throttled submission
      • Appears to have a positive effect
    • MAN: pilot was originaly hardcoded in the pilot; but pilot code still appears to not be fully resolved.
      • lan and wan needs to be set to root (doesn’t matter on the priority).

  • London

    • QMUL drained (deliberately by site)
      • SRM stopped working (connection refused); under investigation

  • SouthGrid

    • OX: 2x10Gb/s bonded links running;
      • Might not be able to try on Xcache unless a new release fixed.

  • Scotgrid

    • NTR

• Other new issues / tasks

 

---

● Ongoing Items

• CentOS7 - Sussex

  • Univa; need to get a user job running.
    • Hopefully enough handles to manipulate PE and pass into RTE

• TPC with http

  • RAL running for ATLAS with Src and Dst

• Storageless Site test (Oxford)

  • NTR

 

---

● News round-table

• Alessandra

  • NTR

• Dan

  • NTR

• Matt

  • Off

• Patrick

  • NTR

• Peter

  • NTR

• Sam

  • NTR

• Vip

  • Ox for downtime next Thurs / Fri

 

10:00 → 10:20 Status

10:20 → 10:40 Ongoing Items

• CentOS7 - Sussex

  • Univa; need to get a user job running.
    • Hopefully enough handles to manipulate PE and pass into RTE

• TPC with http

  • RAL running for ATLAS with Src and Dst

• Storageless Site test (Oxford)

  • NTR

 

10:40 → 10:50 News round-table

• Alessandra

  • NTR

• Dan

  • NTR

• Matt

  • Off

• Patrick

  • NTR

• Peter

  • NTR

• Sam

  • NTR

• Vip

  • Ox for downtime next Thurs / Fri

 

10:50 → 11:00 AOB

ATLAS Monitoring (UK Cloud)