You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-issues@hadoop.apache.org by "Wei-Chiu Chuang (Jira)" <ji...@apache.org> on 2019/10/31 23:42:00 UTC
[jira] [Comment Edited] (HDFS-14794) [SBN read] reportBadBlock is
rejected by Observer.
[ https://issues.apache.org/jira/browse/HDFS-14794?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16964455#comment-16964455 ]
Wei-Chiu Chuang edited comment on HDFS-14794 at 10/31/19 11:41 PM:
-------------------------------------------------------------------
reportBadBlock is initiated by client or DataNode when they detect a block is bad.
And then it is up to the active NameNode to schedule a replica for the bad block / invalidation.
There would be a period of time where observer thinks a bad block is still good. But hopefully the duration is short. Client would retry other replicas IIRC. It's might be okay for observer to process reportBadBlock and mark a block replica corrupt but it should not try to schedule block replication/ invalidation.
In summary, I don't think it affect correctness. Maybe a little drop in availability.
was (Author: jojochuang):
reportBadBlock is initiated by client or DataNode when they detect a block is bad.
And then it is up to the active NameNode to schedule a replica for the bad block / invalidation.
There would be a period of time where observer thinks a bad block is still good. But hopefully the duration is short. Client would retry other replicas IIRC. It's might be okay for observer to process reportBadBlock and mark a block replica corrupt but it should not try to schedule block replication/ invalidation.
> [SBN read] reportBadBlock is rejected by Observer.
> --------------------------------------------------
>
> Key: HDFS-14794
> URL: https://issues.apache.org/jira/browse/HDFS-14794
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: namenode
> Affects Versions: 2.10.0
> Reporter: Konstantin Shvachko
> Priority: Major
>
> {{reportBadBlock}} is rejected by Observer via StandbyException
> {code}StandbyException: Operation category WRITE is not supported in state observer{code}
> We should investigate what are the consequences of this and if we should treat {{reportBadBlock}} as IBRs. Note that {{reportBadBlock}} is a part of both {{ClientProtocol}} and {{DatanodeProtocol}}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org