[jira] [Updated] (HDDS-7252) Polled source Datanodes are wrongly not re-considered for balancing in Container Balancer

["Janus Chow (Jira)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HDDS-7252?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Janus Chow updated HDDS-7252:
-----------------------------
    Target Version/s: 1.5.0  (was: 1.4.0)

I am managing the 1.4.0 release and we currently have more than 500 issues targeted for 1.4.0. I am moving the target field to 1.5.0.
       
If you are actively working on this jira and believe this should be targeted for the 1.4.0 release, Please reach out to me via Apache email or Slack.

> Polled source Datanodes are wrongly not re-considered for balancing in Container Balancer
> -----------------------------------------------------------------------------------------
>
>                 Key: HDDS-7252
>                 URL: https://issues.apache.org/jira/browse/HDDS-7252
>             Project: Apache Ozone
>          Issue Type: Sub-task
>          Components: SCM
>            Reporter: Siddhant Sangwan
>            Priority: Major
>
> The FindSourceGreedy class maintains a priority queue of source datanodes. This class is used to get the next source DN for balancing. According to current design, a datanode can be involved in multiple source->target pairings in one iteration of container balancer.
> When FindSourceGreedy#getNextCandidateSourceDataNode() is called during an iteration, it polls a DN from the queue. This DN is re queued when FindSourceGreedy#increaseSizeLeaving is called. Problem is that if ContainerBalancer#moveContainer returns false, then increaseSizeLeaving will not be called and the DN doesn't get added back.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org