Ceph/CVMFS/Filer Service Meeting

Europe/Zurich
600/R-001 (CERN)

600/R-001

CERN

15
Show room on map
    • 14:00 14:05
      CVMFS 5m
      Speaker: Enrico Bocchi (CERN)
    • 14:05 14:10
      Ceph Upstream News 5m

      Releases, Tickets, Testing, Board, ...

      Speaker: Dan van der Ster (CERN)
    • 14:10 14:15
      Ceph Backends & Block Storage 5m

      Cluster upgrades, capacity changes, rebalancing, ...
      News from OpenStack block storage.

      Speaker: Theofilos Mouratidis (National and Kapodistrian University of Athens (GR))

      ceph/flax:

      • pgs split to 2048 from 1024
      • moved racks RJ35/37 into the default root
      • new racks are ready
      • enabled balancer
    • 14:15 14:20
      Ceph Disk Management 5m

      OSD Replacements, Liaison with CF, Failure Predictions

      Speaker: Julien Collet (CERN)

      Julien:

      • Business as usual: couple of disk replaced
      • Presentation to repair service 8th of may
    • 14:20 14:25
      S3 5m

      Ops, Use-cases (backup, DB), ...

      Speakers: Julien Collet (CERN), Roberto Valverde Cameselle (Universidad de Oviedo (ES))
      • Some rgw vm_kill and no_contact happened due to increased load from atlas RECAST/REANA. Turns out the issue was a failing disk (reallocated sectors) -- the disk was getting super slow but not dead. As a result IOs would pile up in the rgw, consuming memory. TODO:
        • CEPH-712: monitor reallocated sectors
        • CEPH-711: replace our 8GB rgw's with 16GB machines.
          • 5 of which already in production, not yet in use
        • Changed configuration from civetweb frontend to new/better beast. (Supposed to bring performance improvements).
    • 14:25 14:30
      CephFS/HPC/FILER/Manila 5m

      Filer Migration, CephFS/Manila, HPC status and plans.

      Speakers: Dan van der Ster (CERN), Pablo Llopis Sanmillan (CERN)
    • 14:30 14:35
      HyperConverged 5m
      Speakers: Jose Castro Leon (CERN), Julien Collet (CERN), Roberto Valverde Cameselle (Universidad de Oviedo (ES))
      • CDA has been ramping up the Kopano testing with imap and mapi benchmarks (250 users). There is some bottleneck in the kopano code, but ceph volumes and cephfs look almost idle during their testing.
    • 14:35 14:40
      Monitoring 5m

      Julien:

      • Report on ProphetStor experience so far:
        • 3 disks failed in the monitored ros
        • The failed drives all shared a "Abnormal unstable pending sector" conditions
          • We got an email for that

       

      • CEPH-707: lemon to collectd migration in progress

       

    • 14:40 14:45
      AOB 5m