[
https://issues.apache.org/jira/browse/HDDS-5221?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
cchenaxchen updated HDDS-5221:
------------------------------
Description: when three replica were unhealthy,the replication manager will
delete one replica,and the replication would be failed,but this might cause
data loss,so if replication manager find one replica unhealthy,and then
replicate the container from others if at least one replica was healthy,if
not,the replication manager would not send delete command,reserved three
replicas. (was: when three replica were unhealthy,the replication manager will
delete one replica,and the replication would be failed,but this might cause
data loss when datanode)
> when replica is unhealthy,replication manager should replicate first,then
> delete the unhealthy replica
> ------------------------------------------------------------------------------------------------------
>
> Key: HDDS-5221
> URL: https://issues.apache.org/jira/browse/HDDS-5221
> Project: Apache Ozone
> Issue Type: Improvement
> Reporter: cchenaxchen
> Assignee: cchenaxchen
> Priority: Major
>
> when three replica were unhealthy,the replication manager will delete one
> replica,and the replication would be failed,but this might cause data loss,so
> if replication manager find one replica unhealthy,and then replicate the
> container from others if at least one replica was healthy,if not,the
> replication manager would not send delete command,reserved three replicas.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
