[LU-11892] Memory leak in MDT Coordinator Created: 28/Jan/19  Updated: 06/Feb/19  Resolved: 06/Feb/19

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.12.0, Lustre 2.10.6
Fix Version/s: Lustre 2.13.0

Type: Bug Priority: Minor
Reporter: Aurelien Degremont (Inactive) Assignee: Aurelien Degremont (Inactive)
Resolution: Fixed Votes: 0
Labels: hsm

Severity: 3
Rank (Obsolete): 9223372036854775807

 Description   

After creating a big HSM request candidate list, coordinator thread
allocates an update list which is 16 bytes times the number of
candidates files. In some cases, they could be more than 300,000
candidate files and so several MB will be needed.

Memory allocation failure was skipping candidate list freeing which led to memory leak.

 

This problem was detected under Lustre 2.10.5. Since, a patch landed that makes this bug  more difficult to trigger, but the code is still present.

I will submit a fix for it.

 Comments   
Comment by Gerrit Updater [ 28/Jan/19 ]

Aurelien Degremont (degremoa@amazon.com) uploaded a new patch: https://review.whamcloud.com/34123
Subject: LU-11892 hsm: fix memory leak when scheduling HSM requests
Project: fs/lustre-release
Branch: master
Current Patch Set: 1
Commit: d01e04c00b080799001e10b577be2fb15cd0802c

Comment by Gerrit Updater [ 06/Feb/19 ]

Oleg Drokin (green@whamcloud.com) merged in patch https://review.whamcloud.com/34123/
Subject: LU-11892 hsm: fix memory leak when scheduling HSM requests
Project: fs/lustre-release
Branch: master
Current Patch Set:
Commit: fc13209d2a15f68c4b96288b3a68d45826abc64d

Comment by Peter Jones [ 06/Feb/19 ]

Landed for 2.13

Generated at Sat Feb 10 02:47:50 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.