[LU-13390] sanity test_65n: /mnt/lustre/d65n.sanity-4 should show the default layout from /mnt/lustre Created: 26/Mar/20  Updated: 26/Nov/20

Status: Open
Project: Lustre
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Bug Priority: Minor
Reporter: Maloo Assignee: WC Triage
Resolution: Unresolved Votes: 0
Labels: None

Issue Links:
Duplicate
is duplicated by LU-13401 sanity test_65n: /mnt/lustre/d65n.san... Resolved
Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

This issue was created by maloo for Olaf Faaland <faaland1@llnl.gov>

This issue relates to the following test suite run: https://testing.whamcloud.com/test_sets/42273365-55ef-4203-92b3-603f28f2b7b1

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

The test compares the output of "lfs getstripe" for two different directories - the file system root, and subdirectory, and reports they are different when they should be the same.

However, just before the point in the test where "error" is called, without any intervening commands that change filesystem state, lfs getstripe is invoked for each of the two directories and the output is in the test log. The output in the test log does not reflect the difference the error messages indicates.

===
/usr/bin/lfs getstripe -d /mnt/lustre/d65n.sanity-4
stripe_count: 1 stripe_size: 1048576 pattern: raid0 stripe_offset: -1 pool: test_65n

/usr/bin/lfs getstripe -d /mnt/lustre
stripe_count: 1 stripe_size: 1048576 pattern: raid0 stripe_offset: -1 pool: test_65n

sanity test_65n: @@@@@@ FAIL: /mnt/lustre/d65n.sanity-4 should show the default layout from /mnt/lustre
Trace dump:
= /usr/lib64/lustre/tests/test-framework.sh:6142:error()
= /usr/lib64/lustre/tests/sanity.sh:8001:test_65n()
= /usr/lib64/lustre/tests/test-framework.sh:6445:run_one()
= /usr/lib64/lustre/tests/test-framework.sh:6494:run_one_logged()
= /usr/lib64/lustre/tests/test-framework.sh:6334:run_test()
= /usr/lib64/lustre/tests/sanity.sh:8048:main()
===

The actual strings compared by the test are the result of "lfs getstripe" piped into a shell script which attempts to generate the "lfs setstripe" arguments that would produce the same result. So either there is a bug in the shell code that attempts to summarize the lfs getstripe output, or the value returned by lfs getstripe changed after some time passed.

I don't see a bug (or anything non-deterministic) in the shell code.



 Comments   
Comment by Gerrit Updater [ 11/Nov/20 ]

Andreas Dilger (adilger@whamcloud.com) uploaded a new patch: https://review.whamcloud.com/40600
Subject: LU-13390 tests: add debugging to sanity test_65n
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: b2602c251b53df84236f84f5a6c116373355a7db

Comment by Gerrit Updater [ 26/Nov/20 ]

Oleg Drokin (green@whamcloud.com) merged in patch https://review.whamcloud.com/40600/
Subject: LU-13390 tests: add debugging to sanity test_65n
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 039821a5da4e019e921cce5081d4459b02a14954

Generated at Sat Feb 10 03:00:52 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.