[LU-18237] recovery-small: test 10a fails with 'no eviction: before:1726673307'

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Severity: 3
• Assignee: WC Triage
• Reporter: Maloo

Description

This issue was created by maloo for James Simmons <[e-mail]>

This issue relates to the following test suite run: https://testing.whamcloud.com/test_sets/64bcf39a-0c5d-4ebe-aa8e-07c6118719cd

Test session details:
clients: https://build.whamcloud.com/job/lustre-reviews/107661 - 5.14.0-362.24.1.el9_3.x86_64
servers: https://build.whamcloud.com/job/lustre-reviews/107661 - 5.14.0-362.24.1_lustre.el9.x86_64

<<Please provide additional information about the failure here>>

current_state: FULL
state_history:

• [ 1726673245, DISCONN ]
• [ 1726673245, CONNECTING ]
• [ 1726673245, RECOVER ]
• [ 1726673245, FULL ]
• [ 1726673262, DISCONN ]
• [ 1726673262, CONNECTING ]
• [ 1726673262, RECOVER ]
• [ 1726673262, FULL ]
• [ 1726673279, DISCONN ]
• [ 1726673279, CONNECTING ]
• [ 1726673279, RECOVER ]
• [ 1726673279, FULL ]
• [ 1726673297, DISCONN ]
• [ 1726673297, CONNECTING ]
• [ 1726673297, RECOVER ]
• [ 1726673297, FULL ]
  mdc.lustre-MDT0001-mdc-ffff88f406920000.state=
  current_state: FULL
  state_history:
• [ 1726671294, CONNECTING ]
• [ 1726671295, FULL ]
  mdc.lustre-MDT0002-mdc-ffff88f406920000.state=
  current_state: FULL
  state_history:
• [ 1726671294, CONNECTING ]
• [ 1726671295, FULL ]
  mdc.lustre-MDT0003-mdc-ffff88f406920000.state=
  current_state: FULL
  state_history:
• [ 1726671294, CONNECTING ]
• [ 1726671295, FULL ]
  recovery-small test_10a: @@@@@@ FAIL: no eviction: before:1726673307
  Trace dump:
  = /usr/lib64/lustre/tests/test-framework.sh:7177:error()
  = /usr/lib64/lustre/tests/recovery-small.sh:154:test_10a()
  = /usr/lib64/lustre/tests/test-framework.sh:7522:run_one()
  = /usr/lib64/lustre/tests/test-framework.sh:7585:run_one_logged()
  = /usr/lib64/lustre/tests/test-framework.sh:7408:run_test()
  = /usr/lib64/lustre/tests/recovery-small.sh:171:main()
  Dumping lctl log to /autotest/autotest-1/2024-09-18/lustre-reviews_review-dne-part-5_107661_16_4f1e33f4-b320-40c7-8af0-801d1ab9dc56//recovery-small.test_10a.*.1726673420.log
  CMD: trevis-24vm7,trevis-56vm1.trevis.whamcloud.com,trevis-56vm2,trevis-56vm3,trevis-83vm7 /usr/sbin/lctl dk > /autotest/autotest-1/2024-09-18/lustre-reviews_review-dne-part-5_107661_16_4f1e33f4-b320-40c7-8af0-801d1ab9dc56//recovery-small.test_10a.debug_log.\$(hostname -s).1726673420.log;
  dmesg > /autotest/autotest-1/2024-09-18/lustre-reviews_review-dne-part-5_107661_16_4f1e33f4-b320-40c7-8af0-801d1ab9dc56//recovery-small.test_10a.dmesg.\$(hostname -s).1726673420.log
  CMD: trevis-56vm1.trevis.whamcloud.com checkstat -v -p 0777 /mnt/lustre
  /mnt/lustre has perms 0777 OK
  CMD: trevis-83vm7 dmesg
  [ 2314.424588] Lustre: mdt00_001: service thread pid 10192 was inactive for 42.474 seconds. The thread might be hung, or it might only be slow and will resume later. Dumping the stack trace for debugging purposes:
  CMD: trevis-83vm7 dmesg
  [ 2383.543931] Lustre: mdt00_001: service thread pid 10192 completed after 111.597s. This likely indicates the system was overloaded (too many service threads, or not enough hardware resources).

Issue Links

• duplicates: LU-15630 recovery-small test_10a: no eviction: before:1646723217