[
https://issues.apache.org/jira/browse/YARN-634?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13652769#comment-13652769
]
Vinod Kumar Vavilapalli commented on YARN-634:
----------------------------------------------
bq. We don't need YarnRemoteException.getRemoteException* methods at all now?
I meant RPCUtil.getRemoteException* methods
bq. Sure. An unwind method would've been ideal. I'll add this for now, and we
can think of removing it if no longer required when unwind is implemented.
Unwind will be better. We can punt this.
bq. SerializedException is really just a yarn record to captures an error,
which is the payload on most of these FailureEvents. It's part of a method
response payload as against being thrown by the method call itself. So I don't
really see a reason to wind/unwind this at the edge. I'd prefer leaving this as
is, unless you feel strongly about it.
We'll definitely need and unwrap API. Going by the class_name logic, we also
don't need getMessage() and getRemoteTrace() in SerializedException once we
have unwind? I'd rather implement unwrap instead of adding all this unnecessary
functionality.
> Introduce a SerializedException
> -------------------------------
>
> Key: YARN-634
> URL: https://issues.apache.org/jira/browse/YARN-634
> Project: Hadoop YARN
> Issue Type: Sub-task
> Affects Versions: 2.0.4-alpha
> Reporter: Siddharth Seth
> Assignee: Siddharth Seth
> Attachments: YARN-634.txt
>
>
> LocalizationProtocol sends an exception over the wire. This currently uses
> YarnRemoteException. Post YARN-627, this needs to be changed and a new
> serialized exception is required.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
