Michael Smith has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/23186 )

Change subject: IMPALA-14073: Retry when invalidate preempts EP
......................................................................


Patch Set 3:

(3 comments)

http://gerrit.cloudera.org:8080/#/c/23186/3//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/23186/3//COMMIT_MSG@9
PS3, Line 9: IMPALA-13631
> Was the issue already there before IMPALA-13631, so was this issue reproduc
I was able to reproduce some of it with the code changes for IMPALA-13631 
reverted. If this is infrequent enough, I can spend some time reproducing 
exactly how they happen to confirm the different scenarios, but that'll take 
time.


http://gerrit.cloudera.org:8080/#/c/23186/3//COMMIT_MSG@14
PS3, Line 14: Retries eventually succeed, so for now address the test 
instability by
            : retrying on relevant catalog failures.
> What is the goal of the change - to avoid hiding other issues by allowing t
Yes, essentially.


http://gerrit.cloudera.org:8080/#/c/23186/3/tests/custom_cluster/test_concurrent_ddls.py
File tests/custom_cluster/test_concurrent_ddls.py:

http://gerrit.cloudera.org:8080/#/c/23186/3/tests/custom_cluster/test_concurrent_ddls.py@195
PS3, Line 195: if "AnalysisException: Table does not exist: " in err or \
             :        "AnalysisException: Table already exists: " in err or \
             :        "TableNotFoundException: Table not found: " in err:
             :       # This can happen if global invalidate happens while 
renaming table
             :       return True
> Is this pattern specific enough for ALTER TABLE RENAME only?
I think some subsequent commands may fail: rename succeeds, invalidate runs but 
event processor is lagging, next command tries to execute and fails because EP 
is behind.

I need to spend some more time on the exact sequence.



--
To view, visit http://gerrit.cloudera.org:8080/23186
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I73feffec6b059058426701ecc76644704ea0452a
Gerrit-Change-Number: 23186
Gerrit-PatchSet: 3
Gerrit-Owner: Michael Smith <[email protected]>
Gerrit-Reviewer: Csaba Ringhofer <[email protected]>
Gerrit-Reviewer: Fang-Yu Rao <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Michael Smith <[email protected]>
Gerrit-Reviewer: Quanlong Huang <[email protected]>
Gerrit-Reviewer: Riza Suminto <[email protected]>
Gerrit-Comment-Date: Fri, 18 Jul 2025 22:50:52 +0000
Gerrit-HasComments: Yes

Reply via email to