[
https://issues.apache.org/jira/browse/HDFS-7055?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14142925#comment-14142925
]
Colin Patrick McCabe commented on HDFS-7055:
--------------------------------------------
This patch adds tracing to DFSInputStream. It's fairly granulary tracing, done
at the level of read operations. I also added trace spans to the BlockReaders
(both remote and local) so that the amount of time they spend can be
disentangled from other things. I added tracing to {{BlockSender}} so that the
amount of time we spend sending the block can be seen.
By setting {{dfs.client.trace.sampler}}, users can make their HDFS client
create new trace spans on some requests. The available options are
{{NeverSampler}}, {{AlwaysSampler}}, and {{ProbabilitySampler}}.
One thing to keep in mind here is that if you call {{Trace.startSpan}} with
{{Sampler.NEVER}}, and there is an existing thread trace span, a subspan will
always be created. But if there is no existing thread trace span, none will be
created. So there is no need for if statements and null/not null logic here.
You can always get "pass through" behavior by using {{Sampler.NEVER}}.
> Add tracing to DFSInputStream
> -----------------------------
>
> Key: HDFS-7055
> URL: https://issues.apache.org/jira/browse/HDFS-7055
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: datanode, namenode
> Affects Versions: 2.6.0
> Reporter: Colin Patrick McCabe
> Assignee: Colin Patrick McCabe
> Attachments: HDFS-7055.002.patch
>
>
> Add tracing to DFSInputStream.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
