Github user srowen commented on the pull request:
https://github.com/apache/spark/pull/5437#issuecomment-91192902
Referring to the JIRA, I do wonder how much this adds over just setting up
the agent manually with JVM opts, because it touches a reasonable amount of
code. Yes, you have to pick a fixed port, but then problem 2 goes away; you
know what the agent port is. Yes, that doesn't work with several executors per
machine, but if you're restarting your app anyway to debug, would you not
perhaps just run a leaner config with 1 per machine? in the event that you
really have to debug a multiple executor per machine job, maybe you can make
the port a function of the executor's ID? I think we added some support for
that?
---
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]
