Mike Percy has posted comments on this change.

Change subject: KUDU-1330: Add a tool to unsafely recover from loss of majority 
replicas
......................................................................


Patch Set 14:

(6 comments)

http://gerrit.cloudera.org:8080/#/c/6066/14/src/kudu/tools/kudu-admin-test.cc
File src/kudu/tools/kudu-admin-test.cc:

PS14, Line 199: itest:
nit: Can you add a 'using' clause for this and other uses of itest:: stuff, for 
consistency in this file?


PS14, Line 523: 1
Magic number. Why 1? Also, what if the election is churny? Since we know a 
leader got elected I think we should use GetLastOpIdForReplica() on the leader 
to find out what this baseline should be.


PS14, Line 786: 8 
9


PS14, Line 828: 1
Why 1? Same question as above and for the other tests.


PS14, Line 847: >
If we use >= 0 here then we can delete lines 857-859.


PS14, Line 880: 11
Why 11? Shouldn't it be post_commit opid index from line 828 plus 5 for # of 
unsafe config changes plus 1 for the new election no-op?


-- 
To view, visit http://gerrit.cloudera.org:8080/6066
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-MessageType: comment
Gerrit-Change-Id: I908d8c981df74d56dbd034e72001d379fb314700
Gerrit-PatchSet: 14
Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-Owner: Dinesh Bhat <[email protected]>
Gerrit-Reviewer: David Ribeiro Alves <[email protected]>
Gerrit-Reviewer: Dinesh Bhat <[email protected]>
Gerrit-Reviewer: Kudu Jenkins
Gerrit-Reviewer: Mike Percy <[email protected]>
Gerrit-Reviewer: Tidy Bot
Gerrit-HasComments: Yes

Reply via email to