ATLAS UK Cloud Support

Thursday 6 Jan 2022, 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

• • 155460 USER atlas UKI-SOUTHGRID-CAM-HEP less urgent NGI_UK assigned 2022-01-05 11:26:00 Failovers from Cambridge to CERN backup proxy EGI

    • ‘Rogue’ user jobs; admins are contacting users

  • 155430 TEAM atlas UKI-SCOTGRID-ECDF less urgent NGI_UK in progress 2022-01-05 22:18:00 UKI-SCOTGRID-ECDF transfer and deletion errors EGI

    • Major issues with one of the storage nodes, awaiting site prognosis

  • 155410 TEAM atlas RAL-LCG2 less urgent NGI_UK in progress 2022-01-04 07:07:00 RAL-LCG2 jobs failed due to transfer timeout EGI

    • Large number of fts transfers queued; transfers timing out (the time in queue, not the transfer time), leading to compute job failures.
    • Many mitigating steps taken: reduced number of running jobs, new webdav endpoint; dedicated atlas running on two machines on that endpoint.
    • Backlog largely cleared and work ongoing for davs improvements
    • 4-5 GB/s davs reads are sustainable

  • 155141 TEAM atlas UKI-LT2-Brunel less urgent NGI_UK in progress 2021-12-24 08:39:00 Transfers from UKI-LT2-Brunel fail with “Internal Server Error” EGI

    • Aim to resolve tickets shortly. HC test files should have been returned to site (to check); also to check on any outstanding issues

  • 154806 TEAM atlas UKI-LT2-QMUL less urgent NGI_UK in progress 2021-12-25 04:28:00 UKI-LT2-QMUL SOURCE transfer failures: [13] Result (Neon): SSL handshake failed EGI

    • Needs resolving (somehow)

  • 154543 TEAM atlas UKI-SCOTGRID-ECDF urgent NGI_UK in progress 2021-12-08 12:35:00 DPM storage ACL configuration EGI

    • To prod site on how to resolve this

  • 154436 TEAM atlas RAL-LCG2 very urgent NGI_UK on hold 2021-12-08 13:25:00 RAL Echo Davs developments EGI

    • Work in 155410 to help dev work

  • 153367 TEAM atlas RAL-LCG2 urgent NGI_UK on hold 2021-12-01 15:37:00 HTTPS on RAL CTA EGI

    • To be continued …

---

● CPU

• • RAL

    • Poor performance, due to 155410

  • Northgrid

    • Lancs Power issues on Christmas eve; resolved by / on Christmas day

  • London

    • Largely ok; some QMUL blips

  • SouthGrid

    • OK

  • Scotgrid

    • Some issues for Durham; Glasgow CPU efficiency is low

 

---

● Ongoing Items

• TPC with http

  • Active work now restarting for davs deployment

• Storageless Site test (Oxford)

  • Some new Xcache harware may be avaialble for loan
  • Working on finding out why little Xcache traffic
  • Site will upgrade to 5.4.0

• LANCS Storage migration

  • Aliases exist; JW to configure CRIC side

 

 

---

● News round-table

• Gerard

  • NTR

• Matt

  • NTR

• Patrick

  • NTR

• Peter

  • NTR

• Stephen

  • NTR

• Vip

  • NTR

 

 

10:00 → 10:20 Status

10:20 → 10:40 Ongoing Items

• TPC with http

  • Active work now restarting for davs deployment

• Storageless Site test (Oxford)

  • Some new Xcache harware may be avaialble for loan
  • Working on finding out why little Xcache traffic
  • Site will upgrade to 5.4.0

• LANCS Storage migration

  • Aliases exist; JW to configure CRIC side

 

 

10:40 → 10:50 News round-table

• Gerard

  • NTR

• Matt

  • NTR

• Patrick

  • NTR

• Peter

  • NTR

• Stephen

  • NTR

• Vip

  • NTR

 

 

10:50 → 11:00 AOB

ATLAS Monitoring (UK Cloud)