szetszwo commented on PR #5532:
URL: https://github.com/apache/hadoop/pull/5532#issuecomment-1525780573
@umamaheswararao , thanks for reviewing this!
> Could you also check test failures?
The test failures are related. Thanks for filing
szetszwo commented on PR #5532:
URL: https://github.com/apache/hadoop/pull/5532#issuecomment-1516773369
> ... then there should be separate JIRA for it? ...
In the past few days, I was testing whether we need the change below. It
turned that the change lead to a problem that
szetszwo commented on PR #5532:
URL: https://github.com/apache/hadoop/pull/5532#issuecomment-1511669706
> Are we planning to address here? [#5532
(comment)](https://github.com/apache/hadoop/pull/5532#issuecomment-1499923239)
Probably not. I found a potential fix shown below -- it
szetszwo commented on PR #5532:
URL: https://github.com/apache/hadoop/pull/5532#issuecomment-1499923239
> ... when call reaches here, the snapshot id is CURRENT_STATE_ID, which
represent current state. Why are we proceeding to clean subtree originally with
CURRENT_STATE_ID. Could you