Todd Lipcon has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/13664 )

Change subject: IMPALA-7534. Handle invalidation races in CatalogdMetaProvider
......................................................................


Patch Set 3:

(2 comments)

http://gerrit.cloudera.org:8080/#/c/13664/2/fe/src/main/java/org/apache/impala/catalog/local/CatalogdMetaProvider.java
File fe/src/main/java/org/apache/impala/catalog/local/CatalogdMetaProvider.java:

http://gerrit.cloudera.org:8080/#/c/13664/2/fe/src/main/java/org/apache/impala/catalog/local/CatalogdMetaProvider.java@442
PS2, Line 442:     //
> We're not setting hit = true, so we're counting this case as a cache miss?
yea, I was wondering whether to add another separate metric for "piggybacked 
hits" or something, but wasn't sure it was worth it. What do you think?


http://gerrit.cloudera.org:8080/#/c/13664/2/fe/src/main/java/org/apache/impala/catalog/local/CatalogdMetaProvider.java@455
PS2, Line 455:     boolean hit = false;
> Do we have tests for these exceptional code paths? Including the case when
I'm guessing this gets covered by our stress test which does concurrent queries 
and invalidates. I'll double check that this is actually covered and add 
another stress test if not.



--
To view, visit http://gerrit.cloudera.org:8080/13664
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I70f377db88e204825a909389f28dc3451815235c
Gerrit-Change-Number: 13664
Gerrit-PatchSet: 3
Gerrit-Owner: Todd Lipcon <[email protected]>
Gerrit-Reviewer: Bharath Vissapragada <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Tim Armstrong <[email protected]>
Gerrit-Reviewer: Todd Lipcon <[email protected]>
Gerrit-Reviewer: Vihang Karajgaonkar <[email protected]>
Gerrit-Comment-Date: Tue, 18 Jun 2019 20:43:38 +0000
Gerrit-HasComments: Yes

Reply via email to