[LU-3795] replay-single test_58b test_58c: trusted.foo: No such attribute Created: 20/Aug/13  Updated: 21/Oct/13  Resolved: 26/Sep/13

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.5.0
Fix Version/s: Lustre 2.5.0

Type: Bug Priority: Blocker
Reporter: Maloo Assignee: Nathaniel Clark
Resolution: Fixed Votes: 0
Labels: revzfs

Issue Links:
Duplicate
duplicates LU-3669 (mdt_xattr.c:178:mdt_getxattr()) ASSE... Closed
Related
is related to LU-3713 Failure on test suite sanity test_17k... Resolved
is related to LU-2869 extended attribute cache for Lustre Resolved
Severity: 3
Rank (Obsolete): 9813

 Description   

This issue was created by maloo for Nathaniel Clark <nathaniel.l.clark@intel.com>

This issue relates to the following test suite run: http://maloo.whamcloud.com/test_sets/383816e8-097f-11e3-901c-52540035b04c.

The sub-test test_58b failed with the following error:

/mnt/lustre2/d0.replay-single/d58/f.replay-single.58b: trusted.foo: No such attribute
test_58b failed with 1

Info required for matching: replay-single 58b



 Comments   
Comment by Andreas Dilger [ 21/Aug/13 ]

This looks like it also failed on non-zfs testing:
https://maloo.whamcloud.com/sub_tests/749022be-07dd-11e3-bcfa-52540035b04c

and the first failure was on 2013-07-29 00:48:20 UTC
https://maloo.whamcloud.com/sub_tests/e2333076-f87a-11e2-a8b3-52540035b04c

and according to http://review.whamcloud.com/5537 the xattr cache patch landed on 07-29 also.

Comment by Andrew Perepechko [ 21/Aug/13 ]

The first (ZFS) report is a duplicate of LU-3713 which is fixed by http://review.whamcloud.com/7208 in the context of LU-3669.

The second report does not look xattr cache related (no traces of xattr operations either in the MDS or in the client debug logs), "No such file or directory" probably relates to the fact that there was no such file or directory, rather than an xattr failure.

Comment by Peter Jones [ 26/Aug/13 ]

Nathaniel

Please verify whether this is still an issue after LU-3669 fix has landed

Peter

Comment by Nathaniel Clark [ 12/Sep/13 ]

This is verified fixed by LU-3669 closure

Comment by Emoly Liu [ 26/Sep/13 ]

I reopen this ticket because through investigation on LU-3625, I find still many zfs tests hit this error in recent two months, such as:
https://maloo.whamcloud.com/test_sets/b278ad30-24ac-11e3-93a0-52540035b04c
https://maloo.whamcloud.com/test_sets/243e1cd2-24f0-11e3-839b-52540035b04c
https://maloo.whamcloud.com/test_sets/54b1a140-22e3-11e3-9488-52540035b04c
https://maloo.whamcloud.com/test_sets/5bb89698-1a4f-11e3-a95b-52540035b04c

"No such attribute" happens much more frequently than "No such file or directory".

Comment by Andrew Perepechko [ 26/Sep/13 ]

The fixes were landed on Sep, 11 (see http://review.whamcloud.com/#/c/7208/), not two months ago.

It's not clear whether your reports are based on sources more recent than of that date.

https://maloo.whamcloud.com/test_sets/5bb89698-1a4f-11e3-a95b-52540035b04c was ran on Sep 10.
https://maloo.whamcloud.com/test_sets/243e1cd2-24f0-11e3-839b-52540035b04c was ran on Sep 23, the build links to http://review.whamcloud.com/#/c/7584/, which appears to be Patch Set 8 or earlier (Patch set 9 was uploaded on Sep 25) and the parent commit for Patch set 8 is c0e6982b57319e5044cf12bee509c0c571a17cc5, which was landed on Aug 7.

Comment by Emoly Liu [ 26/Sep/13 ]

I searched Maloo test results in this link https://maloo.whamcloud.com/sub_tests/query?utf8=✓&test_set%5Btest_set_script_id%5D=f6a12204-32c3-11e0-a61c-52540025f9ae&sub_test%5Bsub_test_script_id%5D=fb022be0-32c3-11e0-a61c-52540025f9ae&sub_test%5Bstatus%5D=FAIL&sub_test%5Bquery_bugs%5D=&test_session%5Btest_host%5D=&test_session%5Btest_group%5D=&test_session%5Buser_id%5D=&test_session%5Bquery_date%5D=&test_session%5Bquery_recent_period%5D=&test_node%5Bos_type_id%5D=&test_node%5Bdistribution_type_id%5D=&test_node%5Barchitecture_type_id%5D=&test_node%5Bfile_system_type_id%5D=00437f32-318d-11e1-9c6d-5254004bbbd3&test_node%5Blustre_branch_id%5D=&test_node_network%5Bnetwork_type_id%5D=&commit=Update+results

I didn't check each one. Maybe you are right, their parent commit is earlier than LU-3669 fix. If so, please close this ticket and ignore my previous comments.

Comment by Emoly Liu [ 26/Sep/13 ]

Andrew, you are right. Sorry to reopen this one.

Comment by Andrew Perepechko [ 26/Sep/13 ]

Thanks, Emoly. Please, reopen if the problem reoccurs in the future.

Generated at Sat Feb 10 01:36:58 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.