[
https://issues.apache.org/jira/browse/HDFS-9342?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14982192#comment-14982192
]
Walter Su commented on HDFS-9342:
---------------------------------
Both {{updatePipeline}} and {{commitBlock}} will set the visible length of
lastBlock.
It's ok {{commitBlock}} uses {{currentBlockGroup}}, becasue we have flushed all.
{{updatePipeline}} set too bigger visible length according to
{{currentBlockGroup}}, of which the last part can be invisible, since it's not
acked yet.
Thanks [~zhz] for reporting this. I think it is a issue.
> Erasure coding: client should update and commit block based on acknowledged
> size
> --------------------------------------------------------------------------------
>
> Key: HDFS-9342
> URL: https://issues.apache.org/jira/browse/HDFS-9342
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: erasure-coding
> Affects Versions: 3.0.0
> Reporter: Zhe Zhang
> Priority: Critical
>
> For non-EC files, we have:
> {code}
> protected ExtendedBlock block; // its length is number of bytes acked
> {code}
> For EC files, the size of {{DFSStripedOutputStream#currentBlockGroup}} is
> incremented in {{writeChunk}} without waiting for ack. And both
> {{updatePipeline}} and {{commitBlock}} are based on size of
> {{currentBlockGroup}}.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
