[
https://issues.apache.org/jira/browse/HDFS-8172?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14504075#comment-14504075
]
Li Bo commented on HDFS-8172:
-----------------------------
hi, Nicholas
Sorry for the late reply. At first I think we can create a JIRA for a concrete
failure of client writing. If we can handle all kinds of failures under
HDFS-7672, I think that's also very good.
> Failure handling: DFSStripedOutputStream continues writing with enough
> remaining datanodes
> ------------------------------------------------------------------------------------------
>
> Key: HDFS-8172
> URL: https://issues.apache.org/jira/browse/HDFS-8172
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: hdfs-client
> Reporter: Li Bo
> Assignee: Li Bo
>
> For RS(m,n), if no more than n DataStreamers fail, DFSStripedOutputStream
> should continue its writing.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
