Todd Lipcon has posted comments on this change.

Change subject: tablet copy: Allow voting from crashed initial tablet copies
......................................................................


Patch Set 1:

(1 comment)

http://gerrit.cloudera.org:8080/#/c/7961/1//COMMIT_MSG
Commit Message:

PS1, Line 9: The change in c35e4f0093dcdec625d1f647924d854c7bc9f3de missed an
           : important case, which is when the target tablet server in a tablet 
copy
           : operation crashes while in the middle of a tablet copy. In that 
case,
           : the 'tombstone_last_logged_opid' of 1.0 was not persisted in the
           : superblock. This manifested as a very flaky
           : TabletCopyITest.TestTabletCopyNewReplicaFailureCanVote test, which 
would
           : fail about 15% of the time.
can you clarify what effect this has outside of tests? ie is the change you 
referenced just incomplete? (ie it was still an improvement but didn't improve 
all the cases it could have) or was it actually buggy such that it was a 
regression?

put another way, should this be a blocker for the 1.5 release, or is this an 
incremental improvement on top of what's there?


-- 
To view, visit http://gerrit.cloudera.org:8080/7961
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: Idbc9cecac701c24ebe05759adbbb1fbe2b140506
Gerrit-PatchSet: 1
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Mike Percy <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Todd Lipcon <[email protected]>
Gerrit-HasComments: Yes

Reply via email to