Stephen O'Donnell created HDDS-7729:
---------------------------------------
Summary: EC: ECContainerReplicaCount should handle pending delete
of unhealthy replicas
Key: HDDS-7729
URL: https://issues.apache.org/jira/browse/HDDS-7729
Project: Apache Ozone
Issue Type: Sub-task
Components: SCM
Reporter: Jyotirmoy Sinha
Assignee: Stephen O'Donnell
In ECContainerReplicaCount, we intentionally ignore UNHEALTHY replicas so that
when we check for over or under replication, it is if the replicas are not
there.
However if there is a pending delete for an unhealthy replica, due to ignoring
the unhealthy replicas, the pending delete decrements the healthy count and
makes the container appear under replicated when it is not.
When processing pending deletes, we need also ignore any that are removing the
ignored unhealthy replicas.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
