Uploaded image for project: 'Data Foundation Bugs'
  1. Data Foundation Bugs
  2. DFBUGS-730

[2321538] OSD pods on CLBO state in ODF 4.18 due to segmentation fault On Power

XMLWordPrintable

    • Icon: Bug Bug
    • Resolution: Unresolved
    • Icon: Critical Critical
    • odf-4.18
    • odf-4.18
    • ceph/RADOS/ppc64
    • None
    • True
    • Hide

      Install is blocked

      Show
      Install is blocked
    • False
    • ?
    • ppc64le
    • ?
    • If docs needed, set a value
    • Proposed
    • None

      Description of problem:

      OSD pods are going in CLBO state after installation of ODF 4.18 on ppc64le
      Seeing below logs on OSD pods

      ```
      debug 2024-10-24T16:32:56.740+0000 7fff84b20020 0 bdev(0x1000d810e00 /var/lib/ceph/osd/ceph-0/block) ioctl(F_SET_FILE_RW_HINT) on /var/lib/ceph/osd/ceph-0/block failed: (22) Invalid argument
      debug 2024-10-24T16:32:56.740+0000 7fff84b20020 1 bdev(0x1000d810e00 /var/lib/ceph/osd/ceph-0/block) open backing device/file reports st_blksize 65536, using bdev_block_size 4096 anyway
      debug 2024-10-24T16:32:56.740+0000 7fff84b20020 1 bdev(0x1000d810e00 /var/lib/ceph/osd/ceph-0/block) open size 536870912000 (0x7d00000000, 500 GiB) block_size 4096 (4 KiB) rotational device, discard not supported
      debug 2024-10-24T16:32:56.740+0000 7fff84b20020 1 bluestore(/var/lib/ceph/osd/ceph-0) _set_cache_sizes cache_size 3221225472 meta 0.45 kv 0.45 kv_onode 0.04 data 0.06
      debug 2024-10-24T16:32:56.740+0000 7fff84b20020 1 bdev(0x1000d810e00 /var/lib/ceph/osd/ceph-0/block) close
      debug 2024-10-24T16:32:57.100+0000 7fff84b20020 1 mClockScheduler: set_osd_capacity_params_from_config: osd_bandwidth_cost_per_io: 499321.90 bytes/io, osd_bandwidth_capacity_per_shard 19660800.00 bytes/second
      debug 2024-10-24T16:32:57.100+0000 7fff84b20020 0 osd.0:7.OSDShard using op scheduler mclock_scheduler, cutoff=196
      --: line 14: 932 Segmentation fault (core dumped) "${@}"
      + wait 932
      + ceph_osd_rc=139
      + '[' 139 -eq 0 ']'
      + exit 139
      ```

      Version-Release number of selected component (if applicable):
      OCP: 4.18.0-ec.2
      ODF: 4.18.0-41

      Steps to Reproduce:
      1. Install OCP 4.18 and ODF 4.18.0-41 with local storage

      Actual results:
      OSD pods are in CLBO state

      Expected results:
      OSD pods should be running

        1. ceph-osd-failed.tar.gz
          198.15 MB
          Sudeesh John
        2. core.ceph-osd.167.cc10f2d31b64498cb12b09241eb74223.95746.1730179655000000.zst
          12.82 MB
          Sudeesh John
        3. segfault in container IBM Z.txt
          52 kB
          Aliaksei Makarau
        4. valgrind.txt
          3.68 MB
          Sudeesh John

              rh_sjohn Sudeesh John
              rh-ee-shdas Shilpi Das
              Elad Ben Aharon Elad Ben Aharon
              Votes:
              0 Vote for this issue
              Watchers:
              14 Start watching this issue

                Created:
                Updated: