[LU-9764] recovery-double-scale_pairwise_fail test failed: mount.lustre: mount /dev/vdb at /mnt/mds3 failed: Bad file descriptor

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.12.0
• Severity: 3
• Assignee: nasf (Inactive)
• Reporter: nasf (Inactive)

Description

It is reported that:

Starting mds3: -o rw,user_xattr  /dev/vdb /mnt/mds3
lm0413: mount.lustre: increased /sys/block/vdb/queue/max_sectors_kb from 1024 to 2147483647
lm0413: mount.lustre: mount /dev/vdb at /mnt/mds3 failed: Bad file descriptor
pdsh@lm0417: lm0413: ssh exited with exit code 9
Start of /dev/vdb on mds3 failed 9
 recovery-double-scale test_pairwise_fail: @@@@@@ FAIL: Restart of mds3 failed!

  Trace dump:
  = /usr/lib64/lustre/tests/test-framework.sh:4976:error()
  = /usr/lib64/lustre/tests/test-framework.sh:1232:mount_facets()
  = /usr/lib64/lustre/tests/test-framework.sh:2583:facet_failover()
  = /usr/lib64/lustre/tests/recovery-double-scale.sh:61:reboot_recover_node()
  = /usr/lib64/lustre/tests/recovery-double-scale.sh:145:failover_pair()
  = /usr/lib64/lustre/tests/recovery-double-scale.sh:247:test_pairwise_fail()
  = /usr/lib64/lustre/tests/test-framework.sh:5236:run_one()
  = /usr/lib64/lustre/tests/test-framework.sh:5274:run_one_logged()
  = /usr/lib64/lustre/tests/test-framework.sh:5078:run_test()
  = /usr/lib64/lustre/tests/recovery-double-scale.sh:303:main()
Dumping lctl log to /tmp/test_logs/1496261766/recovery-double-scale.test_pairwise_fail.*.1496262136.log
lm0411: Warning: Permanently added 'lm0417,192.168.4.17' (ECDSA) to the list of known hosts.
lm0412: Warning: Permanently added 'lm0417,192.168.4.17' (ECDSA) to the list of known hosts.
lm0415: Warning: Permanently added 'lm0417,192.168.4.17' (ECDSA) to the list of known hosts.
lm0418: Warning: Permanently added 'lm0417,192.168.4.17' (ECDSA) to the list of known hosts.
lm0420: Warning: Permanently added 'lm0417,192.168.4.17' (ECDSA) to the list of known hosts.
lm0413: Warning: Permanently added 'lm0417,192.168.4.17' (ECDSA) to the list of known hosts.
lm0419: Warning: Permanently added 'lm0417,192.168.4.17' (ECDSA) to the list of known hosts.

The log shows that:

[ 653.209594] LustreError: 11853:0:(lfsck_namespace.c:6525:lfsck_namespace_setup()) lustre-MDT0002-osd: fail to init namespace LFSCK component: rc = -9 [ 653.215672] LustreError: 11853:0:(mdd_device.c:1065:mdd_prepare()) lustre-MDD0002: failed to initialize lfsck: rc = -9 [ 653.219548] LustreError: 11853:0:(obd_mount_server.c:1834:server_fill_super()) Unable to start targets: -9 [ 653.224335] Lustre: Failing over lustre-MDT0002 [ 657.283581] Lustre: server umount lustre-MDT0002 complete [ 657.286706] LustreError: 11853:0:(obd_mount.c:1445:lustre_fill_super()) Unable to mount (-9) [ 657.507398] Lustre: DEBUG MARKER: recovery-double-scale test_pairwise_fail: @@@@@@ FAIL: Restart of mds3 failed!

Issue Links

• is related to: LU-10027 Unable to finish mount on MDS while Rolling downgrade