[
https://issues.apache.org/jira/browse/HDFS-16999?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17719619#comment-17719619
]
ASF GitHub Bot commented on HDFS-16999:
---------------------------------------
Hexiaoqiao commented on PR #5622:
URL: https://github.com/apache/hadoop/pull/5622#issuecomment-1535670070
@zhangshuyan0 Thanks for your proposal and try to fix this issue. Just
glance the PR, it propose to switch `processFirstBlockReport` to
`processReport` when restart DataNode only, right? I am concerned the
performance if we do that.
a. Is it possible to improve `processFirstBlockReport` to solve this issue?
b. Will it affect the process logic or performance when add one new DataNode
to cluster?
Thanks again.
> Fix wrong use of processFirstBlockReport()
> ------------------------------------------
>
> Key: HDFS-16999
> URL: https://issues.apache.org/jira/browse/HDFS-16999
> Project: Hadoop HDFS
> Issue Type: Bug
> Reporter: Shuyan Zhang
> Assignee: Shuyan Zhang
> Priority: Major
> Labels: pull-request-available
>
> `processFirstBlockReport()` is used to process first block report from
> datanode. It does not calculating `toRemove` list because it believes that
> there is no metadata about the datanode in the namenode. However, If a
> datanode is re registered after restarting, its `blockReportCount` will be
> updated to 0. That is to say, the first block report after a datanode
> restarts will be processed by `processFirstBlockReport()`. This is
> unreasonable because the metadata of the datanode already exists in namenode
> at this time, and if redundant replica metadata is not removed in time, the
> blocks with insufficient replicas cannot be reconstruct in time, which
> increases the risk of missing block. In summary, `processFirstBlockReport()`
> should only be used when the namenode restarts, not when the datanode
> restarts.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
