[
https://issues.apache.org/jira/browse/HDDS-6596?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17535837#comment-17535837
]
Kaijie Chen commented on HDDS-6596:
-----------------------------------
Yes, I have done this before based on HDDS-6518 in our internal fork.
We can get List<BlockData> from each DataNode, and then grouping them by the
LocalID.
We still need index of each DN in the EC pipeline to form the blocks to a EC
stripe.
And each BlockData contains List<ChunkInfo>, which can be used in ReadChunk
later.
To make ListBlock rpc easier to use, the change proposed in HDDS-6520 can be
applied.
> EC: Get list of blocks in CoordinatorDN
> ---------------------------------------
>
> Key: HDDS-6596
> URL: https://issues.apache.org/jira/browse/HDDS-6596
> Project: Apache Ozone
> Issue Type: Sub-task
> Reporter: Kaijie Chen
> Assignee: Kaijie Chen
> Priority: Major
>
> When CoordinatorDN receives `RecoverECContainer` from SCM,
> it should first list all blocks in the container group to find the missing
> ones.
>
> Full design doc by [~markgui]:
> https://docs.google.com/document/d/1dM-qOYuXJ2Yehj_NfIaOullFbun5h3S1nDLsQi1KjqM/edit
--
This message was sent by Atlassian Jira
(v8.20.7#820007)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
