[
https://issues.apache.org/jira/browse/HDDS-16664?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ivan Andika updated HDDS-16664:
-------------------------------
Summary: ReplicationManager should use the storage type from datanode
command (was: ReplicationManager should use the storage type from datandoe
command)
> ReplicationManager should use the storage type from datanode command
> --------------------------------------------------------------------
>
> Key: HDDS-16664
> URL: https://issues.apache.org/jira/browse/HDDS-16664
> Project: Apache Ozone
> Issue Type: Sub-task
> Reporter: Ivan Andika
> Assignee: Ivan Andika
> Priority: Major
>
> Currently, ReplicationManager#adjustPendingOpsAndMetrics ignore both
> ECReconstructionTarget.getStorageType() and
> ReplicateContainerCommand.getTargetVolumeStorageType() and use the container
> storage type instead.
> If an SSD container falls back to DISK, SCM reserves SSD while the command
> writes DISK; another repair can then consume the same final DISK slot.
> Accounting should use each command target’s StorageType, including null for
> an unconstrained target.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
