[LU-7791] signal vs CP callback race

• Type: Bug
• Resolution: Fixed
• Priority: Minor
• Fix Version/s: Lustre 2.12.7, Lustre 2.12.8
• Labels: LTS12 patch
• Severity: 3
• Assignee: Andriy Skulysh
• Reporter: Andriy Skulysh

Description

a lock isn't canceled on the server on reception of a signal, thus client is evicted for not responding to LDLM_FL_AST_SENT.

Issue Links

• is related to: LU-8067 sanityn test_31a: read dd: reading `/mnt/lustre2/d31a.sanityn/f31a.sanityn': Input/output error 0 blocks, must be

Comments

[Gerrit Updater]

"Oleg Drokin <[e-mail]>" merged in patch https://review.whamcloud.com/44297/
Subject: LU-7791 ldlm: signal vs CP callback race
Project: fs/lustre-release
Branch: b2_12
Current Patch Set:
Commit: 8f4336481a482544f54b7e8a9d63340a57062d68

[Andreas Dilger]

Oleg, I saw that the last time this patch was backported to b2_12, it was later reverted due to LU-8067, linked here:

commit 51d9874087a87a9a616615adc98f6239ebd8619d
Author:     Oleg Drokin <[e-mail]>
AuthorDate: Tue Apr 26 17:46:09 2016 +0000
CommitDate: Mon May 2 17:41:30 2016 +0000

    Revert "LU-7791 ldlm: signal vs CP callback race"
    
    This causes frequent failures tracked in LU-8067
    
    This reverts commit a880f385164547837a34df92acd6cdc8e9a9698d.
    
    Change-Id: Ib247bb924dabe4635edd17f9a2dca391135f558f
    Reviewed-on: http://review.whamcloud.com/19798

[Gerrit Updater]

Minh Diep ([e-mail]) uploaded a new patch: https://review.whamcloud.com/44297
Subject: LU-7791 ldlm: signal vs CP callback race
Project: fs/lustre-release
Branch: b2_12
Current Patch Set: 1
Commit: 59eb3b254804ace64503863e9305fd4532cfb662

[Peter Jones]

Landed for 2.14

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch https://review.whamcloud.com/19898/
Subject: LU-7791 ldlm: signal vs CP callback race
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 7fff052c930da4822c3b2a13d130da7473a20a58

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/19798/
Subject: Revert "LU-7791 ldlm: signal vs CP callback race"
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 51d9874087a87a9a616615adc98f6239ebd8619d

[Gerrit Updater]

Andriy Skulysh ([e-mail]) uploaded a new patch: http://review.whamcloud.com/19898
Subject: LU-7791 ldlm: signal vs CP callback race
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 41adbd3e8c6832a5d883a74e5b89ce35acc5b43b

[Andreas Dilger]

Reopen ticket because patch was reverted due to frequent sanityn test_31a failures (LU-8067).

Please update and resubmit whole patch with fix.

[Gerrit Updater]

Oleg Drokin ([e-mail]) uploaded a new patch: http://review.whamcloud.com/19798
Subject: Revert "LU-7791 ldlm: signal vs CP callback race"
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: d632c523656693320d5838d720374087d9d47cdf

[Joseph Gmitter (Inactive)]

Landed to master for 2.9.0