Adar Dembo has posted comments on this change.

Change subject: master: fix corruption when AlterTable() races with 
CreateTable()
......................................................................


Patch Set 1:

(2 comments)

The test failure was in the initial write to ITScannerMultiTablet. I suspect 
it's flakiness, but I don't know for sure because kudu-gerrit wasn't archiving 
failsafe-based test logs. I've reconfigured it to do that and will keep an eye 
out going forward.

http://gerrit.cloudera.org:8080/#/c/3607/1/src/kudu/master/catalog_manager.cc
File src/kudu/master/catalog_manager.cc:

Line 880:       SetupError(resp->mutable_error(), 
MasterErrorPB::TABLE_NOT_FOUND, s);
> While we're revisiting this code-- is TABLE_NOT_FOUND really the right erro
See https://gerrit.cloudera.org/#/c/2714 for why this combination was chosen. I 
agree that it's somewhat unintuitive, but I don't think there's a slam dunk 
option amongst any of our existing error codes.


http://gerrit.cloudera.org:8080/#/c/3607/1/src/kudu/master/master-test.cc
File src/kudu/master/master-test.cc:

Line 1087:         CreateTableRequestPB req;
> Why not use CreateTable here?
I need access to resp.error().code(); CreateTable only gives me  
resp.error().status(). Figured this was the path of least resistance, as in 
TestConcurrentCreateOfSameTable.


-- 
To view, visit http://gerrit.cloudera.org:8080/3607
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I6c9e4214c09bc47a5a10b12d6ffe8b35906708c9
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Adar Dembo <a...@cloudera.com>
Gerrit-Reviewer: Adar Dembo <a...@cloudera.com>
Gerrit-Reviewer: Dan Burkert <d...@cloudera.com>
Gerrit-Reviewer: David Ribeiro Alves <dral...@apache.org>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <t...@apache.org>
Gerrit-HasComments: Yes

Reply via email to