Github user tgravescs commented on a diff in the pull request:
https://github.com/apache/spark/pull/5719#discussion_r29337481
--- Diff: docs/running-on-yarn.md ---
@@ -134,6 +134,13 @@ Most of the configs are the same for Spark on YARN as
for other deployment modes
</td>
</tr>
<tr>
+ <td><code>spark.yarn.am.port</code></td>
+ <td>(random)</td>
+ <td>
+ Port for the YARN Application Master to listen on. This is used for
communicating with the executors.
--- End diff --
this is not the channel used to communicate with executors.
This channel is used in yarn client mode to communicate between the spark
driver running on a gateway and the application master (which is doing
container requests) running on yarn.
It also looks like I was wrong my previous comment. It actually is also
used in yarn cluster mode. Its used for the dynamic executor feature, where it
is handling the kill from the scheduler backend.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
