Ashwani Raina has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/24476 )

Change subject: KUDU-3767 Add global leader awareness to auto leader rebalancer 
(Part 2)
......................................................................


Patch Set 3:

(1 comment)

http://gerrit.cloudera.org:8080/#/c/24476/1//COMMIT_MSG
Commit Message:

http://gerrit.cloudera.org:8080/#/c/24476/1//COMMIT_MSG@7
PS1, Line 7: Add global leader awareness to auto leader rebalancer (Part 2)
> Thanks for the questions!
Thank you for the responses.
My understanding and assumption of the logic aligns with what you have stated 
here.

However, in reality, I found somewhat different behavior.
I want to point out that in my example the number of tables is > 1.
I tried a straightforward use case with 2 tables with 9 tablets each and RF of 
3. After enabling auto_leader_rebalancing_enabled and waiting for about half an 
hour, with no workload running , there seems to be no change in number of 
leaders. The global state of leader replica distribution is not even:

Tablet Server Summary
               UUID               |    Address     | Status  | Location | 
Tablet Leaders | Active Scanners
----------------------------------+----------------+---------+----------+----------------+-----------------
 24aabf171a834916b4d47e0261b60896 | 127.0.0.1:9874 | HEALTHY | <none>   |      
10        |       0
 a2bc89ee4c4a44029c9c3c08267225df | 127.0.0.1:9872 | HEALTHY | <none>   |       
4        |       0
 bcd18e79b65f4a2eabeb841323b3292b | 127.0.0.1:9870 | HEALTHY | <none>   |       
4        |       0
---
For one table 'default.loadgen_auto_a0d0bdec2a5743cf8b8b992b3ab589dc':
tserver(9874) was hosting 6 leader replicas
tserver(9872) was hosting 1 leader replica
tserver(9870) was hosting 2 leader replicas

---
I restarted the cluster with your changes but there was no change in leadership 
counts, even after almost 30 minutes.

I would expect every tserver to host 6 leader replicas each and for each table, 
there should be 3 leader replicas per tserver. But none of these seem to be 
happening.

FYI, I used 'perf loadgen' Kudu CLI option along with 
'table_num_range_partitions' and 'table_num_replicas' to create tables with 
desired number of partitions and replica factor.

Am I missing something?



--
To view, visit http://gerrit.cloudera.org:8080/24476
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: kudu
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I5d44761f6444ca1d1bf4fcee405d65d029e41980
Gerrit-Change-Number: 24476
Gerrit-PatchSet: 3
Gerrit-Owner: Gabriella Lotz <[email protected]>
Gerrit-Reviewer: Alexey Serbin <[email protected]>
Gerrit-Reviewer: Ashwani Raina <[email protected]>
Gerrit-Reviewer: Gabriella Lotz <[email protected]>
Gerrit-Reviewer: Kudu Jenkins (120)
Gerrit-Reviewer: Marton Greber <[email protected]>
Gerrit-Reviewer: Zoltan Chovan <[email protected]>
Gerrit-Comment-Date: Wed, 08 Jul 2026 13:26:42 +0000
Gerrit-HasComments: Yes

Reply via email to