jiajunwang opened a new issue #660: Controller leader switch may trigger some 
additional (unexpected) rebalance in the WAGED rebalancer
URL: https://github.com/apache/helix/issues/660
 
 
   1. Since the WAGED rebalancer determines the rebalance scope by comparing 
in-memory cluster data cache, it is possible that a controller leader switch 
impacts this scope calculating. Then the scope could be enlarged or shrank.
   2. In addition, if the leader switch happened while a rebalance pipeline is 
proceeding, we might see 2 versions output of the best possible state from the 
different controller (need to confirm if the pipeline would continue once the 
leadership switch happens. If that's the case, some partitions might be moved 
back and forth.
   
   To solve this,
   1. Considering reset the data change detector on new session handling.
   2. We need to confirm that a rebalance pipeline will be stopped once the 
leadership of the controller has been switched.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to