[
https://issues.apache.org/jira/browse/HDDS-6539?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17517629#comment-17517629
]
Uma Maheswara Rao G commented on HDDS-6539:
-------------------------------------------
In our initial discussions, we thought this can create lot of edge case while
doing the offline recovery. SO, to avoid the design complexity in offline
recovery we did make stripe fail and create new block group if any failure.
Since this is only req for smaller cluster, providing thing support might not
justify to take the design complexity. I am closing this until you have clear
thoughts on offline recovery complexities with this support.
Feel free to re-open when you have some design thoughts around that.
> EC: write when the datanode not enough
> --------------------------------------
>
> Key: HDDS-6539
> URL: https://issues.apache.org/jira/browse/HDDS-6539
> Project: Apache Ozone
> Issue Type: Sub-task
> Reporter: cchenaxchen
> Assignee: cchenaxchen
> Priority: Major
>
> when the ec use 3+2, when datanode only have 4, so the ec write will fail, we
> should guarantee the ec write usability when datanode number less than 5.
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
