[LU-1750] Update next_id even the previous process already clear the orphans Created: 15/Aug/12  Updated: 26/Feb/13  Resolved: 26/Feb/13

Status: Resolved
Project: Lustre
Component/s: None
Affects Version/s: Lustre 2.4.0
Fix Version/s: None

Type: Bug Priority: Blocker
Reporter: Di Wang Assignee: Di Wang
Resolution: Won't Fix Votes: 0
Labels: None

Severity: 3
Rank (Obsolete): 3984

 Description   

During orphan cleanup, OST should always update last_id, even though the previous threads might already cleanup orphans. Considering following case

1. mds start, send clear orphan req to OST, and OST begin to clear orphan, but then MDS fails before it gets reply. i.e. it does not update next_id yet.
2. mds restart and send clear orphan req again to OST with original next_id, then it found the orphan has been cleanup, so still reply with the original id, which is wrong, and we should update next_id in this case.



 Comments   
Comment by Di Wang [ 15/Aug/12 ]

http://review.whamcloud.com/3679

Comment by Di Wang [ 26/Feb/13 ]

The patch is actually for obdfilter, but we do not use obdfilter anymore in 2.4. The related fix has been exists in OFD.

Generated at Sat Feb 10 01:19:21 UTC 2024 using Jira 9.4.14#940014-sha1:734e6822bbf0d45eff9af51f82432957f73aa32c.