Sumit Agrawal created HDDS-13617:
------------------------------------
Summary: Avoid immediate ICR for close container
Key: HDDS-13617
URL: https://issues.apache.org/jira/browse/HDDS-13617
Project: Apache Ozone
Issue Type: Sub-task
Components: Ozone Datanode
Reporter: Sumit Agrawal
ICR is being send for every close container event from SCM. This is send even
there is no state change.
Number of ICR increases too much when DN pipeline state becomes un-stable due
to other DNs going down, network issue OR this datanode being very busy. This
is in result of pipeline close which creates close container event for all
containers in pipeline.
And this gets retried.
To avoid un-necessary load, close container report can be notified over next
HB, not an immediate triger.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
