Csaba Ringhofer has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/21029 )

Change subject: IMPALA-12543: Detect self-events before finishing DDL
......................................................................


Patch Set 10:

(7 comments)

http://gerrit.cloudera.org:8080/#/c/21029/10//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/21029/10//COMMIT_MSG@11
PS10, Line 11: happens
I think that this is a bit misleading - the issue could probably happen on all 
build types, just it was witnessed on these ones, right?


http://gerrit.cloudera.org:8080/#/c/21029/10//COMMIT_MSG@15
PS10, Line 15: locks is
nit: lock is


http://gerrit.cloudera.org:8080/#/c/21029/10/fe/src/main/java/org/apache/impala/service/CatalogOpExecutor.java
File fe/src/main/java/org/apache/impala/service/CatalogOpExecutor.java:

http://gerrit.cloudera.org:8080/#/c/21029/10/fe/src/main/java/org/apache/impala/service/CatalogOpExecutor.java@1078
PS10, Line 1078:     public void cancelInflightEvent() {
Can we call this when inflightEventRegistered_ is false? If not, then a 
precondition could be added for this.


http://gerrit.cloudera.org:8080/#/c/21029/10/fe/src/main/java/org/apache/impala/service/CatalogOpExecutor.java@1081
PS10, Line 1081:       LOG.info("Cancel in-progress in-flight event of table " 
+ table_.getFullName()
               :           + ". isInsertEvent=" + isInsertEvent_ + " 
versionNumber=" + newVersionNumber_
               :           + " isRemoved=" + isRemoved);
Can you add a comment about why isRemoved can be false?


http://gerrit.cloudera.org:8080/#/c/21029/10/fe/src/main/java/org/apache/impala/service/CatalogOpExecutor.java@1390
PS10, Line 1390:       tbl.resetInProgressModification();
Shouldn't we do something with `modification` in case of an exception? Can we 
assume that modification.hasInProgressModification() is false at this point?


http://gerrit.cloudera.org:8080/#/c/21029/10/fe/src/main/java/org/apache/impala/service/CatalogOpExecutor.java@6447
PS10, Line 6447: TException
This won't catch exceptions from alterTableWithTransaction as it wraps the 
TException as ImpalaRuntimeException. The old try catch could be kept to around 
the non-transactional case to do this wrappin and the cancelInflightEvent could 
be called in an outer catch block


http://gerrit.cloudera.org:8080/#/c/21029/10/fe/src/main/java/org/apache/impala/service/CatalogOpExecutor.java@6452
PS10, Line 6452: {
This could be done outside the try block at line 6424
+nit: no braces needed



--
To view, visit http://gerrit.cloudera.org:8080/21029
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I8365c934349ad21a4d9327fc11594d2fc3445f79
Gerrit-Change-Number: 21029
Gerrit-PatchSet: 10
Gerrit-Owner: Riza Suminto <[email protected]>
Gerrit-Reviewer: Csaba Ringhofer <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Jason Fehr <[email protected]>
Gerrit-Reviewer: Quanlong Huang <[email protected]>
Gerrit-Reviewer: Riza Suminto <[email protected]>
Gerrit-Reviewer: Sai Hemanth Gantasala <[email protected]>
Gerrit-Comment-Date: Wed, 06 Mar 2024 16:02:51 +0000
Gerrit-HasComments: Yes

Reply via email to