You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ozone.apache.org by "Sammi Chen (Jira)" <ji...@apache.org> on 2021/08/30 08:48:00 UTC
[jira] [Updated] (HDDS-5697) SCMContainerPlacementRackAware should
both affinity nodes before fallback
[ https://issues.apache.org/jira/browse/HDDS-5697?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Sammi Chen updated HDDS-5697:
-----------------------------
Summary: SCMContainerPlacementRackAware should both affinity nodes before fallback (was: SCMContainerPlacementRackAware should try to find node on the same rack as second affinity node before fallback)
> SCMContainerPlacementRackAware should both affinity nodes before fallback
> -------------------------------------------------------------------------
>
> Key: HDDS-5697
> URL: https://issues.apache.org/jira/browse/HDDS-5697
> Project: Apache Ozone
> Issue Type: Bug
> Reporter: Sammi Chen
> Assignee: Sammi Chen
> Priority: Major
>
> If an under replicated container has 2 replicas, SCMContainerPlacementRackAware only chooses the node on the rack of first replica, if no available node found, it just fallback to ignore the rack topology. In the case, SCMContainerPlacementRackAware should try the second replica too before it fallback.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@ozone.apache.org
For additional commands, e-mail: issues-help@ozone.apache.org