[
https://issues.apache.org/jira/browse/HDDS-1888?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Nanda kumar updated HDDS-1888:
------------------------------
Resolution: Fixed
Fix Version/s: 0.5.0
0.4.1
Status: Resolved (was: Patch Available)
> Add containers to node2container map in SCM as part of ICR processing
> ---------------------------------------------------------------------
>
> Key: HDDS-1888
> URL: https://issues.apache.org/jira/browse/HDDS-1888
> Project: Hadoop Distributed Data Store
> Issue Type: Bug
> Components: SCM
> Reporter: Nanda kumar
> Assignee: Nanda kumar
> Priority: Major
> Labels: pull-request-available
> Fix For: 0.4.1, 0.5.0
>
> Time Spent: 1h 50m
> Remaining Estimate: 0h
>
> In SCM node2container and node2pipeline maps are managed by NodeManager and
> pipeline2container map is managed by PipelineManager.
> Currently, when a container is allocated in SCM, it is added to
> pipeline2container map and we are not adding it to node2container map. We
> update the node2container map only when the datanode sends full container
> report.
> When a node is marked as dead, DeadNodeHandler processes the event and it
> gets the list of containers that are hosted by the dead datanode and updates
> the respective container replica state in ContainerManager. The list of
> containers on the datanode is read from node2container map, this map will be
> missing containers which are created recently (after the last container
> report). In such cases we not be able to remove the container replica
> information for those containers. In reality, these containers are under
> replicated, but SCM will never know.
> We should add containers to node2container map in SCM as soon as a container
> is allocated.
--
This message was sent by Atlassian JIRA
(v7.6.14#76016)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
