[LU-16682] sanity-pfl test_1c: comp4 stripe count != 2000

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

Description

This issue was created by maloo for S Buisson <[e-mail]>

This issue relates to the following test suite run: https://testing.whamcloud.com/test_sets/4808a017-c473-4019-91e0-c790c3c97661

test_1c failed with the following error:

comp4 stripe count 1397 != 2000

Test session details:
clients: https://build.whamcloud.com/job/lustre-reviews/93380 - 4.18.0-372.32.1.el8_6.x86_64
servers: https://build.whamcloud.com/job/lustre-reviews/93380 - 4.18.0-372.32.1.el8_lustre.x86_64

VVVVVVV DO NOT REMOVE LINES BELOW, Added by Maloo for auto-association VVVVVVV
sanity-pfl test_1c - comp4 stripe count 1397 != 2000

Issue Links

• is related to: LU-11912 reduce number of OST objects created per MDS Sequence

Comments

[Gerrit Updater]

"Andreas Dilger <[e-mail]>" uploaded a new patch: https://review.whamcloud.com/c/fs/lustre-release/+/50531
Subject: LU-16682 tests: verify OSTSEQWIDTH fix
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 7197e2c295dbb32bb17903b476463c57ec79dd78

[Andreas Dilger]

It looks like there is already a patch in LU-11912.

[Andreas Dilger]

Hmm, it looks like it has failed for several different patches starting on 2023-03-29, and not in the 4 weeks before that, so likely a regression due to a patch. The failures are for patches that haven't landed yet, so it is unlikely that they are the cause.

Patches landed on this day:

a7222127c7 LU-16642 tests: improve sanity-sec test_61
8f40a3d711 LU-16639 misc: cleanup concole messages
e998d21caf LU-16589 tests: add sanity/31l to test ln command
17bbf5bdd6 LU-930 docs: fix whatis output
36cbba150b LU-16632 tests: more margin of error for sanity/56xh
91a3726f31 LU-16633 obdclass: fix rpc slot leakage
12c3465199 LU-14291 batch: don't include lustre_update.h for client only builds
d5b26443a3 LU-16615 utils: add messages in l_getidentity
b30f825232 LU-16601 kernel: update SLES15 SP4 [5.14.21-150400.24.46.1]
8f004bc53b LU-16599 obdclass: job_stats can parse escaped jobid string
fc7a0d6013 LU-14668 lnet: add 'lock_prim_nid" lnet module parameter
f5293fb66e LU-16598 osp: cleanup comment in osp_syn
c.c
5e24b374f7 LU-16595 test: save one second in wait_destroy_complete()
da230373bd LU-16563 lnet: use discovered ni status to set initial health
0366422cfd LU-16221 kernel: update RHEL 9.1 [5.14.0-162.18.1.el9_1]
2d40d96b4e LU-15053 tests: reset quota if ENABLE_QUOTA=1
7e893c7095 LU-16382 build: udev files in /usr/lib
b33808d3ae LU-16338 readahead: clip readahead with kms
ccee6b92ec LU-13107 utils: remove duplicate lctl erase/fork_lcfg
2471d35c0e LU-16217 iokit: Add lst.sh wrapper and lst-survey
bdbc7f9f42 LU-12805 tests: disable replay-single/36
73ee638813 LU-16604 kfilnd: kfilnd_peer ref leak on send
6fab1fe4a5 LU-9680 lnet: handle multi-rail setups
0ecb2a167c LU-11912 ofd: reduce LUSTRE_DATA_SEQ_MAX_WIDTH
c97d4cdf4d LU-16629 osd: refill the existing env

At first glance I would suspect LU-11912 as the culprit, since it is changing OST object allocation to stress test that code in autotest (this would be much less of an issue in production).

[Andreas Dilger]

Sebastien, did this happen on a filesystem with fscrypt enabled? I'm wondering if it failed due to a lack of xattr space for holding 2000 stripes, or if it is just a race condition? I haven't seen it before.