[LU-16872] sanity: test_27M Error: '(5) stripe count , should be 8 for append' - Whamcloud Community JIRA

Details

Type: Bug
Resolution: Fixed
Priority: Critical
Fix Version/s: Lustre 2.16.0
Affects Version/s: None
Labels:
None

Severity:
3
Rank (Obsolete):
9223372036854775807

Description

This issue was created by maloo for Frank Sehr <fsehr@whamcloud.com>

This issue relates to the following test suite run: https://testing.whamcloud.com/test_sets/46d1b101-8e4f-4415-bb49-ee39963275fe

Test session details:
clients: https://build.whamcloud.com/job/lustre-reviews/95319 - 4.18.0-425.10.1.el8_7.x86_64
servers: https://build.whamcloud.com/job/lustre-reviews/95319 - 4.18.0-425.10.1.el8_lustre.x86_64

== sanity test 27M: test O_APPEND striping =============== 03:40:09 (1685936409)
CMD: trevis-129vm4 /usr/sbin/lctl get_param -n version 2>/dev/null
striped dir -i3 -c2 -H crush2 /mnt/lustre/d27M.sanity
CMD: trevis-129vm4 /usr/sbin/lctl get_param -n mdd.lustre-MDT0000.append_pool
CMD: trevis-129vm4 /usr/sbin/lctl get_param -n mdd.lustre-MDT0000.append_stripe_count
CMD: trevis-129vm4,trevis-129vm5 /usr/sbin/lctl set_param mdd.*.append_stripe_count=0
mdd.lustre-MDT0000.append_stripe_count=0
mdd.lustre-MDT0002.append_stripe_count=0
mdd.lustre-MDT0001.append_stripe_count=0
mdd.lustre-MDT0003.append_stripe_count=0
CMD: trevis-129vm4,trevis-129vm5 /usr/sbin/lctl set_param mdd.*.append_stripe_count=2
mdd.lustre-MDT0000.append_stripe_count=2
mdd.lustre-MDT0002.append_stripe_count=2
mdd.lustre-MDT0001.append_stripe_count=2
mdd.lustre-MDT0003.append_stripe_count=2
CMD: trevis-129vm4,trevis-129vm5 /usr/sbin/lctl set_param mdd.*.append_stripe_count=-1
mdd.lustre-MDT0000.append_stripe_count=-1
mdd.lustre-MDT0002.append_stripe_count=-1
mdd.lustre-MDT0001.append_stripe_count=-1
mdd.lustre-MDT0003.append_stripe_count=-1
/usr/lib64/lustre/tests/sanity.sh: line 3101: /mnt/lustre/d27M.sanity/f27M.sanity.5: Invalid argument
lfs: getstripe for '/mnt/lustre/d27M.sanity/f27M.sanity.5' failed: No such file or directory
/usr/lib64/lustre/tests/sanity.sh: line 3103: [: -eq: unary operator expected
sanity test_27M: @@@@@@ FAIL: (5) stripe count , should be 8 for append
Trace dump:
= /usr/lib64/lustre/tests/test-framework.sh:6585:error()
= /usr/lib64/lustre/tests/sanity.sh:3104:test_27M()
= /usr/lib64/lustre/tests/test-framework.sh:6925:run_one()
= /usr/lib64/lustre/tests/test-framework.sh:6974:run_one_logged()
= /usr/lib64/lustre/tests/test-framework.sh:6811:run_test()
= /usr/lib64/lustre/tests/sanity.sh:3181:main()
Dumping lctl log to /autotest/autotest-2/2023-06-05/lustre-reviews_review-ldiskfs-dne_95319_27_26b27a74-2421-4453-9c33-cd237feca413//sanity.test_27M.*.1685936415.log
CMD: trevis-129vm1.trevis.whamcloud.com,trevis-129vm2,trevis-129vm3,trevis-129vm4,trevis-129vm5 /usr/sbin/lctl dk > /autotest/autotest-2/2023-06-05/lustre-reviews_review-ldiskfs-dne_95319_27_26b27a74-2421-4453-9c33-cd237feca413//sanity.test_27M.debug_log.$(hostname -s).1685936415.log;
dmesg > /autotest/autotest-2/2023-06-05/lustre-reviews_review-ldiskfs-dne_95319_27_26b27a74-2421-4453-9c33-cd237feca413//sanity.test_27M.dmesg.$(hostname -s).1685936415.log
CMD: trevis-129vm4,trevis-129vm5 /usr/sbin/lctl set_param mdd.*.append_stripe_count=1
mdd.lustre-MDT0000.append_stripe_count=1
mdd.lustre-MDT0002.append_stripe_count=1
mdd.lustre-MDT0001.append_stripe_count=1
mdd.lustre-MDT0003.append_stripe_count=1
CMD: trevis-129vm4,trevis-129vm5 /usr/sbin/lctl set_param mdd.*.append_pool=none
mdd.lustre-MDT0000.append_pool=none
mdd.lustre-MDT0002.append_pool=none
mdd.lustre-MDT0001.append_pool=none
mdd.lustre-MDT0003.append_pool=none

Attachments

Issue Links

duplicates

LU-16014 sanity test_27M: crash in lod_qos_prep_create()

Resolved

is duplicated by

LU-16147 sanity test_27M: "stripe count , should be 8 for append"

Resolved

LU-16648 sanity test_27M: crashed in lod_statfs_and_check()

Resolved

is related to

LU-16648 sanity test_27M: crashed in lod_statfs_and_check()

Resolved

is related to

LU-11279 sanity test_65c: lverify failed

Resolved

LU-15727 lod_get_default_lov_striping() misinterprets composite striping for append

Resolved

(1 is related to )

Activity

[LU-16872] sanity: test_27M Error: '(5) stripe count , should be 8 for append'

Peter Jones added a comment - 19/Aug/23 12:05 PM

Landed for 2.16

Peter Jones added a comment - 19/Aug/23 12:05 PM Landed for 2.16

Gerrit Updater added a comment - 19/Aug/23 5:33 AM

"Oleg Drokin <green@whamcloud.com>" merged in patch https://review.whamcloud.com/c/fs/lustre-release/+/51602/
Subject: ~~LU-16872~~ tests: exercise sanity test_27M more fully
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 7bb1685048bf999df03ceadab39faa09b8a5560d

Gerrit Updater added a comment - 19/Aug/23 5:33 AM "Oleg Drokin <green@whamcloud.com>" merged in patch https://review.whamcloud.com/c/fs/lustre-release/+/51602/ Subject: LU-16872 tests: exercise sanity test_27M more fully Project: fs/lustre-release Branch: master Current Patch Set: Commit: 7bb1685048bf999df03ceadab39faa09b8a5560d

Gerrit Updater added a comment - 19/Aug/23 5:33 AM

"Oleg Drokin <green@whamcloud.com>" merged in patch https://review.whamcloud.com/c/fs/lustre-release/+/51559/
Subject: ~~LU-16872~~ lod: reset llc_ostlist when using O_APPEND stripes
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 766b35a9700f36aa08b652fa9d18b890d34bf4a5

Gerrit Updater added a comment - 19/Aug/23 5:33 AM "Oleg Drokin <green@whamcloud.com>" merged in patch https://review.whamcloud.com/c/fs/lustre-release/+/51559/ Subject: LU-16872 lod: reset llc_ostlist when using O_APPEND stripes Project: fs/lustre-release Branch: master Current Patch Set: Commit: 766b35a9700f36aa08b652fa9d18b890d34bf4a5

Gerrit Updater added a comment - 07/Jul/23 8:05 PM

"Andreas Dilger <adilger@whamcloud.com>" uploaded a new patch: https://review.whamcloud.com/c/fs/lustre-release/+/51602
Subject: ~~LU-16872~~ tests: exercise sanity test_27M more fully
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 221a2d05d5d4ec2b39c88c6a5d84df2ba3f177dc

Gerrit Updater added a comment - 07/Jul/23 8:05 PM "Andreas Dilger <adilger@whamcloud.com>" uploaded a new patch: https://review.whamcloud.com/c/fs/lustre-release/+/51602 Subject: LU-16872 tests: exercise sanity test_27M more fully Project: fs/lustre-release Branch: master Current Patch Set: 1 Commit: 221a2d05d5d4ec2b39c88c6a5d84df2ba3f177dc

Thomas Bertschinger added a comment - 04/Jul/23 2:08 PM

How to reproduce:

# setup:
$ lctl set_param mdd.*.append_stripe_count=-1
$ lfs setstripe -o 1,3 /mnt/lustre

# touch enough files with the default striping so that every mdt kernel thread probably has the defaults stored in its memory
$ for i in {0..100}; do touch /mnt/lustre/x$i; done

# now an append should return EINVAL as long as it gets handled by a kernel thread that previously did a create with default stripes
$ echo 1 >> /mnt/lustre/f
-bash: /mnt/lustre/g: Invalid argument

A closely related problem occurs when an append_pool is set, but in this case, the create succeeds, but the append file is created with the default stripes rather than the pool.

I haven't identified which patch caused (or uncovered) the issue yet – I didn't see anything obvious in the patches merged shortly before the first test failure occurred. So I'll attempt a git bisect to try to find what caused this and will update if I get that answer.

Thomas Bertschinger added a comment - 04/Jul/23 2:08 PM How to reproduce: # setup: $ lctl set_param mdd.*.append_stripe_count=-1 $ lfs setstripe -o 1,3 /mnt/lustre # touch enough files with the default striping so that every mdt kernel thread probably has the defaults stored in its memory $ for i in {0..100}; do touch /mnt/lustre/x $i ; done # now an append should return EINVAL as long as it gets handled by a kernel thread that previously did a create with default stripes $ echo 1 >> /mnt/lustre/f -bash: /mnt/lustre/g: Invalid argument A closely related problem occurs when an append_pool is set, but in this case, the create succeeds, but the append file is created with the default stripes rather than the pool. I haven't identified which patch caused (or uncovered) the issue yet – I didn't see anything obvious in the patches merged shortly before the first test failure occurred. So I'll attempt a git bisect to try to find what caused this and will update if I get that answer.

Gerrit Updater added a comment - 04/Jul/23 1:56 PM

"Thomas Bertschinger <bertschinger@lanl.gov>" uploaded a new patch: https://review.whamcloud.com/c/fs/lustre-release/+/51559
Subject: ~~LU-16872~~ lod: do not stripe O_APPEND files on specific OSTs
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: 81e2439da705dd35dc2c8c687be21cf7dc952eba

Gerrit Updater added a comment - 04/Jul/23 1:56 PM "Thomas Bertschinger <bertschinger@lanl.gov>" uploaded a new patch: https://review.whamcloud.com/c/fs/lustre-release/+/51559 Subject: LU-16872 lod: do not stripe O_APPEND files on specific OSTs Project: fs/lustre-release Branch: master Current Patch Set: 1 Commit: 81e2439da705dd35dc2c8c687be21cf7dc952eba

Thomas Bertschinger added a comment - 29/Jun/23 6:51 PM

Oops, I wasn't thinking straight this morning but overflowing llc_pool wouldn't affect op_array and op_count since the buffer wouldn't be adjacent to these fields. So I'm still looking for what could cause op_array and op_count to have bad values. (I still think that's the most likely explanation for the issue.)

Thomas Bertschinger added a comment - 29/Jun/23 6:51 PM Oops, I wasn't thinking straight this morning but overflowing llc_pool wouldn't affect op_array and op_count since the buffer wouldn't be adjacent to these fields. So I'm still looking for what could cause op_array and op_count to have bad values. (I still think that's the most likely explanation for the issue.)

People

Assignee:: Thomas Bertschinger

Reporter:: Maloo

Votes:: 0 Vote for this issue

Watchers:: 8 Start watching this issue

Dates

Created:: 05/Jun/23 5:19 PM

Updated:: 16/Jan/24 9:34 PM

Resolved:: 19/Aug/23 12:05 PM