[LU-3455] Test failure replay-single, test_20b: after 23936 > before 9600

• Type: Bug
• Resolution: Fixed
• Priority: Critical
• Fix Version/s: Lustre 2.7.0, Lustre 2.5.4
• Affects Version/s: Lustre 2.5.0, Lustre 2.6.0, Lustre 2.5.2, Lustre 2.5.3
• Labels: HB prz zfs
• Severity: 3
• Assignee: Hongchao Zhang
• Reporter: Maloo

Description

This issue was created by maloo for Nathaniel Clark <[e-mail]>

This issue relates to the following test suite run:
http://maloo.whamcloud.com/test_sets/27d96f00-bf41-11e2-88e0-52540035b04c
https://maloo.whamcloud.com/test_sets/96ebdda0-c483-11e2-bc10-52540035b04c
https://maloo.whamcloud.com/test_sets/7a22e7fe-c762-11e2-ae6b-52540035b04c
https://maloo.whamcloud.com/test_sets/911e66d4-c520-11e2-9b87-52540035b04c
https://maloo.whamcloud.com/test_sets/e4570cf4-c4d3-11e2-ac71-52540035b04c

The sub-test test_20b failed with the following error:

after 23936 > before 9600

Info required for matching: replay-single 20b

Issue Links

• is related to: LU-3626 replay-single test_20b: @@@@@@ FAIL: test_20b failed with 3
• is related to: LU-4051 Timeout in replay-single, subtest test_20b

Comments

[nasf (Inactive)]

We still hit it on b2_5 with ZFS backend:
https://testing.hpdd.intel.com/test_sets/05e4fbe4-bd23-11e4-a946-5254006e85c2

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/13231/
Subject: LU-3455 mdt: sync when evicting client
Project: fs/lustre-release
Branch: b2_5
Current Patch Set:
Commit: fd7d369b7a513fd0b3123d16b8329a0cfc97875c

[Jodi Levi (Inactive)]

Patch landed to Master. b2_5 patch is being tracked externally.

[Gerrit Updater]

Jian Yu ([e-mail]) uploaded a new patch: http://review.whamcloud.com/13231
Subject: LU-3455 mdt: sync when evicting client
Project: fs/lustre-release
Branch: b2_5
Current Patch Set: 1
Commit: aa79c4974653d101be9d03200914b023970a7d41

[Gerrit Updater]

Oleg Drokin ([e-mail]) merged in patch http://review.whamcloud.com/11767/
Subject: LU-3455 mdt: sync when evicting client
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: fabd01968d0fac83e4855da5a5dfe8f68c038ceb

[Isaac Huang (Inactive)]

This might be related to LU-4265. Free space accounting with ZFS can be quite tricky as sometimes frees can be delayed for quite a while.

[nasf (Inactive)]

Another failure instance:
https://testing.hpdd.intel.com/test_sets/d89c27d6-5b1e-11e4-9c62-5254006e85c2

[Hongchao Zhang]

the patch has been updated to sync the underlying OSD if some client is being evicted.

[Hongchao Zhang]

the initial patch is tracked at http://review.whamcloud.com/#/c/11767/

[Jian Yu]

Lustre Build: https://build.hpdd.intel.com/job/lustre-b2_5/86/ (2.5.3 RC1)
FSTYPE=zfs

The same failure occurred: https://testing.hpdd.intel.com/test_sets/158df844-311d-11e4-b503-5254006e85c2