[
https://issues.apache.org/jira/browse/HDFS-14968?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17016082#comment-17016082
]
Ahmed Hussein commented on HDFS-14968:
--------------------------------------
JUnit tests in the report seem to fail intermittently:
# {{TestRedudantBlocks}}: HDFS-15092
# {{TestBalancerRPCDelay}}: HDFS-14757
# {{TestDecommission.testAllocAndIBRWhileDecommission}}: I wil file a new Jira
to fix it. It is more likely to be a race condition reading the block state
without waiting for it to be come {{COMPLETE}}.
[~kihwal], Can you please take a look at the patch?
> Add ability to know datanode staleness
> --------------------------------------
>
> Key: HDFS-14968
> URL: https://issues.apache.org/jira/browse/HDFS-14968
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: datanode, logging, namenode
> Reporter: Ahmed Hussein
> Assignee: Ahmed Hussein
> Priority: Minor
> Attachments: HDFS-14968.001.patch, HDFS-14968.002.patch,
> HDFS-14968.003.patch
>
>
> There is no way to know whether a DataNode was marked stale or no longer
> stale by the NameNode.
> It will be good to have the option to enable logging the DataNode staleness
> to figure out if the staleness was the reason behind remote reads.
> Therefore, analyze performance and decision making of the local vs remote
> reads.
>
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
