Stephen O'Donnell created HDDS-2593:
---------------------------------------
Summary: DatanodeAdminMonitor should track unreplicated containers
and complete the admin workflow accordingly
Key: HDDS-2593
URL: https://issues.apache.org/jira/browse/HDDS-2593
Project: Hadoop Distributed Data Store
Issue Type: Sub-task
Components: SCM
Affects Versions: 0.5.0
Reporter: Stephen O'Donnell
Assignee: Stephen O'Donnell
HDDS-2459 allowed the replicationManager to take care of containers which are
under-replicated due to decommission and maintenance.
Its also exposed a new API to return a ContainerReplicaCount object:
{code}
getContainerReplicaCount(Container container)
{code}
This object will allow the DatanodeAdminMonitor to check if each container is
"sufficiently replicated" before decommission or maintenance can complete and
hence can be used to track the progress of each node as it progresses though
the admin workflow.
We should track the containers on each node in administration and ensure that
each is closed and sufficiently replicated before allowing decommission to
complete.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
