[
https://issues.apache.org/jira/browse/SPARK-15708?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15583751#comment-15583751
]
Alex Bozarth commented on SPARK-15708:
--------------------------------------
I'm not sure closing this as cannot reproduce was correct, but I'm not sure how
it could be fixed either. Due to the nature of those tables they get the host
string from entirely different places in code. For the task table it's stored
in {{TaskInfo}} but for the Agg. Metrics tables it's stored in
{{BlockManagerId}}. The better question is when can these two end up with
different host strings (IP vs hostname) and why. [~tgraves] is this something
you would want fixed or was it just an behavioral oddity?
> Tasks table in Detailed Stage page shows ip instead of hostname under
> Executor ID/Host
> --------------------------------------------------------------------------------------
>
> Key: SPARK-15708
> URL: https://issues.apache.org/jira/browse/SPARK-15708
> Project: Spark
> Issue Type: Bug
> Components: Web UI
> Affects Versions: 2.0.0
> Reporter: Thomas Graves
> Priority: Minor
>
> If you go to the detailed Stages page in Spark 2.0, the Tasks table under the
> Executor ID/Host columns hosts the hostname as an ip address rather then a
> fully qualified hostname.
> The table above it (Aggregated Metrics by Executor) shows the "Address" as
> the full hostname.
> I'm running spark on yarn on latest branch-2.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
