Batch Operations Weekly

Europe/Zurich
31-S-023 (CERN)

31-S-023

CERN

22
Show room on map
Videoconference Rooms
Batch_Operations_Weekly
Name
Batch_Operations_Weekly
Description
Batchops
Extension
10748105
Owner
Gavin McCance
Auto-join URL
Useful links
Phone numbers
    • 14:00 15:00
      Agenda 1h

      Reminders

      • Discuss new fillup | capacity planner | you name it
        • terraform with a data source to provide the count?

      Agenda

      • Job sandbox & job removal / eviction (rota ticket XYZ)

        • Do we want to put jobs on HOLD rather than delete when their time expires.
        • Document: behaviour & benchmark thing
        • Benchmark jobs to be put on HOLD
      • disktype project metadata request to cloud

        • Send iointensive jobs to ssd
        • "legacy" projects have this set to "hdd"
      • [BBC-2239] ATLAS T0 to drain / rebuild to HTCondor + CC7

        • Trying for half now / half next week
      • GPU servers ready for installation to batch
      • Migrate AMS to t0 cell?

        • Fixes different quotas issue
        • Can use for backfill
      • Static slot backfill

        • Need separate collector for static slots
        • Monitoring changes needed to query backfill slots
      • [BBC-2109] CC7 Migration:

        • Going to 64% by the end of the month.
      • LHCb reporting “ghost” jobs.

        • Hunting for low efficiency jobs
      • Haggis:

        • Deploying the website behind a reverse proxy:
          • No need to worry about certificates. Life is great!
          • Website <-> HTTP <-> Reverse Proxy <-> HTTPS <-> OAuth2 provider (CERN)
        • CRUD prototype deployed on https://haggis-website-govue-01.cern.ch (Resourcestab only)
          • Two possibilities on how to implement CRUD for Accounting Groups:
            • Another tab for Accounting Groups that follows the same logic employed on the Resources tab
            • Be able to create Accounting Groups from the Resources tab (<— this)
              • Need to think about the UX aspect of it
      • Terraform:

        • WIP merge requests for batch-infrastructure. Let me know of missing bits and pieces.
        • Enabling TLS: checking correctness for requesting certificates on k8s minions with an alias.
        • Recipe available for prometheus/metrics on k8s (not yet in upstream).
      • Round Table:
        • flavors with swap being added, puppet patch to not add /pool/swap
Your browser is out of date!

Update your browser to view this website correctly. Update my browser now

×