Alexey Serbin has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/23909 )

Change subject: KUDU-3728 Add range-aware rebalancing to auto_rebalancer
......................................................................


Patch Set 6:

(1 comment)

http://gerrit.cloudera.org:8080/#/c/23909/6/src/kudu/master/auto_rebalancer.cc
File src/kudu/master/auto_rebalancer.cc:

http://gerrit.cloudera.org:8080/#/c/23909/6/src/kudu/master/auto_rebalancer.cc@617
PS6, Line 617: static_cast<uint16_t>(static_cast<uint8_t>(range_key_begin[i])
> Yes, that's right, the ksck and auto-rebalancer should not diverge. I used
I'd think of keeping things consistent in the Kudu git source repository and 
binaries.  And I also think it's a good idea to address the signed extension 
issue in ksck (apparently, that seem to be unexpected given the formatting for 
the std::ostringstream object, so we can call it a bug).   Meanwhile, it's 
better to separate concerns into two different changelists/patches.

With that, I'd think of the following two steps:

1) In this particular patch, make the code in auto_rebalancer using the same 
convention as in ksck (feel free to add a TODO or mention a JIRA issue, see 
below)

2) Address the signed extension issue, correcting the code in both ksck and 
here in a separate patch.  If it makes sense (at least, for easier tracking of 
this for users and developers), consider filing a ticket in the Apache Kudu 
upstream JIRA and tag the changelist with the JIRA ticket.

Thanks a lot!



--
To view, visit http://gerrit.cloudera.org:8080/23909
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Idf716bef8c21395da98dd7c9e33001190d9b5587
Gerrit-Change-Number: 23909
Gerrit-PatchSet: 6
Gerrit-Owner: Gabriella Lotz <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Gabriella Lotz <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Comment-Date: Wed, 04 Mar 2026 19:11:14 +0000
Gerrit-HasComments: Yes

Reply via email to