[
https://issues.apache.org/jira/browse/CASSANDRA-21033?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Isaac Reath updated CASSANDRA-21033:
------------------------------------
Test and Documentation Plan: Created dtest to reproduce the issue.
Status: Patch Available (was: Open)
> DiskUsageBroadcaster does not update usageInfo on node replacement
> ------------------------------------------------------------------
>
> Key: CASSANDRA-21033
> URL: https://issues.apache.org/jira/browse/CASSANDRA-21033
> Project: Apache Cassandra
> Issue Type: Bug
> Components: Feature/Guardrails, Transactional Cluster Metadata
> Reporter: Isaac Reath
> Assignee: Isaac Reath
> Priority: Normal
> Fix For: 6.x
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> While working on CASSANDRA-21024, I found that on trunk we are not calling
> the `onRemove` method for IEndpointStateChangeSubscriber implementations when
> a node is replaced. This leads to the {{DiskUsageBroadcaster}} to not remove
> the node which was replaced, which makes calls to
> {{DiskUsageBroadcaster#hasStuffedOrFullNode}} to incorrectly return true if a
> {{STUFFED}} / {{FULL}} node is replaced with a {{SPACIOUS}} node.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
