[GitHub] [hadoop] Hexiaoqiao commented on pull request #5583: HDFS-16987. [BugFix] MarkBlockAsCorrupt should not mark a replica as corrupted if the DN has a newest replica

["Hexiaoqiao (via GitHub)" <gi...@apache.org>]

Hexiaoqiao commented on PR #5583:
URL: https://github.com/apache/hadoop/pull/5583#issuecomment-1535684657

   @ZanderXu thanks for your detailed comments. IIUC, the replica's GS is monotonically increasing at DataNode side, right? If that I think we could improve it rely on the basic rule at NameNode side,
   a. ignore one replica with older GS at the same DataNode.
   b. mark corrupt when one replica with same GS NameNode maintained but older GS compare to other replicas.
   If this is true, improve `markBlockAsCorrupt` could fix this issue. Please correct me if I missing some information. Thanks.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org


---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: common-issues-help@hadoop.apache.org