Bharath Vissapragada has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/11638 )

Change subject: IMPALA-7669: Gracefully handle concurrent invalidate/partial 
fetch RPCs
......................................................................


Patch Set 2:

(3 comments)

http://gerrit.cloudera.org:8080/#/c/11638/2//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/11638/2//COMMIT_MSG@22
PS2, Line 22: tight loop in parallel shells
> worthwhile making an e2e test for this? we have a couple of these for query
IIRC I added a few sleeps in some codepaths to repro this, since the race 
window was small. Lemme try to write an e-e test and see if it repros without 
the patch.


http://gerrit.cloudera.org:8080/#/c/11638/2/fe/src/main/java/org/apache/impala/catalog/IncompleteTable.java
File fe/src/main/java/org/apache/impala/catalog/IncompleteTable.java:

http://gerrit.cloudera.org:8080/#/c/11638/2/fe/src/main/java/org/apache/impala/catalog/IncompleteTable.java@145
PS2, Line 145: cause_ == null
> This could be guarded equivalently at the call-site with isLoaded(). Here i
you mean use !isLoaded() here instead of (cause_ == null)?


http://gerrit.cloudera.org:8080/#/c/11638/2/fe/src/main/java/org/apache/impala/catalog/IncompleteTable.java@149
PS2, Line 149: more meaningful lookup status
> perhaps TABLE_NOT_LOADED?
will do.



--
To view, visit http://gerrit.cloudera.org:8080/11638
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I8533f73f25ca42a20f146ddfd95d4213add9b705
Gerrit-Change-Number: 11638
Gerrit-PatchSet: 2
Gerrit-Owner: Bharath Vissapragada <[email protected]>
Gerrit-Reviewer: Bharath Vissapragada <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Todd Lipcon <[email protected]>
Gerrit-Reviewer: Vuk Ercegovac <[email protected]>
Gerrit-Comment-Date: Tue, 16 Oct 2018 17:44:48 +0000
Gerrit-HasComments: Yes

Reply via email to