Stephen O'Donnell created HDDS-7727:
---------------------------------------
Summary: EC: SCM unregistered event handler for
DatanodeCommandCountUpdated
Key: HDDS-7727
URL: https://issues.apache.org/jira/browse/HDDS-7727
Project: Apache Ozone
Issue Type: Sub-task
Components: SCM
Reporter: Stephen O'Donnell
Assignee: Attila Doroszlai
In SCM I see logs like:
{code}
2023-01-04 23:56:45,346 WARN org.apache.hadoop.hdds.server.events.EventQueue:
No event handler registered for event TypedEvent{payloadType=DatanodeDetails,
name='Datanode_Command_Queue_Updated'}
{code}
It seems an event handler was not registered correctly for the new events which
are getting produced, which causes noise in the logs and perhaps the messages
to build up in the queue and may result in memory issues.
This event was created to notify RM when the queue is updated via a heartbeat
so it can adjust replication load limits. The handler
`DatanodeCommandCountUpdatedHandler` was for this purpose, but perhaps it is
not wired up correctly.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
