[LU-8279] sanity-scrub test_4b: @@@@@@ FAIL: Error in dmesg detected Created: 15/Jun/16  Updated: 04/May/17  Resolved: 17/Jun/16

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.9.0
Fix Version/s: Lustre 2.9.0

Type: Bug Priority: Blocker
Reporter: Maloo Assignee: nasf (Inactive)
Resolution: Fixed Votes: 0
Labels: None

Issue Links:
Related
is related to LU-9433 sanity-scrub test_6: Error in dmesg d... Resolved
is related to LU-8278 sanity-scrub test_4b: Error in dmesg ... Resolved
Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

This issue was created by maloo for nasf <fan.yong@intel.com>

Please provide additional information about the failure here.

This issue relates to the following test suite run: https://testing.hpdd.intel.com/test_sets/e3ebda8e-3257-11e6-80b9-5254006e85c2.

There is inode reference leak

CMD: trevis-15vm1.trevis.hpdd.intel.com,trevis-15vm2,trevis-15vm3,trevis-15vm4,trevis-15vm8 dmesg
Kernel error detected: [11602.132484] VFS: Busy inodes after unmount of dm-1. Self-destruct in 5 seconds.  Have a nice day...
[11616.134774] VFS: Busy inodes after unmount of dm-3. Self-destruct in 5 seconds.  Have a nice day...
[11610.008080] VFS: Busy inodes after unmount of dm-2. Self-destruct in 5 seconds.  Have a nice day...
 sanity-scrub test_4b: @@@@@@ FAIL: Error in dmesg detected 


 Comments   
Comment by Gerrit Updater [ 15/Jun/16 ]

Fan Yong (fan.yong@intel.com) uploaded a new patch: http://review.whamcloud.com/20792
Subject: LU-8279 scrub: fix inode reference leak
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: c47ef8662201217076bac09dcf50c867e7752545

Comment by Andreas Dilger [ 15/Jun/16 ]

Is there any explanation why this started failing? Is it due to some patch being landed that had a bug? If yes, did that patch also fail testing, but this problem was missed?

Comment by Andreas Dilger [ 15/Jun/16 ]

It looks like this is related to the landing of one of the patches:

It would be worthwhile to investigate if this problem hit on either of those patches before landing?

Comment by nasf (Inactive) [ 15/Jun/16 ]

It is introduced by the patch http://review.whamcloud.com/#/c/16951/. But related failure under such patch has never been triggered before.

Comment by nasf (Inactive) [ 15/Jun/16 ]

That may be related with the system schedule order: if the OI scrub run faster, and it could has already repaired the invalid OI mapping for the remote directory object before related RPC handler to call osd_fid_lookup() to locate such remote directory object.

Comment by Gerrit Updater [ 16/Jun/16 ]

Oleg Drokin (oleg.drokin@intel.com) merged in patch http://review.whamcloud.com/20792/
Subject: LU-8279 scrub: fix inode reference leak
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: 1738f00d4a4cee01a2c39a6313acabdcb5775269

Comment by nasf (Inactive) [ 17/Jun/16 ]

The patch has been landed to master.

Generated at Sat Feb 10 02:16:09 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.