[
https://issues.apache.org/jira/browse/HDFS-7055?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14158254#comment-14158254
]
Colin Patrick McCabe commented on HDFS-7055:
--------------------------------------------
bq. I agree but still see 280 spans on .005 patch.
If you trace every read (sampler set to Sampler.ALWAYS), then yes, you will see
a lot of spans. If you want a smaller number of spans, using the
{{ProbabilitySampler}} would be a better choice.
Most of the spans shown in that picture are {{DFSInputStream#byteArrayRead}}.
But this is something that needs to be traced when tracing is on. It's one of
the entry points to the HDFS read path. If we don't trace this entry point, we
have no way of associating any trace spans underneath with the HDFS read that
caused them. In the context of HBase, this would also mean we didn't know
where HBase's spans ended and HDFS's began.
Let's get this in and then we can continue tweaking trace points.
> 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, HDFS-7055.003.patch,
> HDFS-7055.004.patch, HDFS-7055.005.patch, screenshot-get-1mb.005.png,
> screenshot-get-1mb.png
>
>
> Add tracing to DFSInputStream.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
