[
https://issues.apache.org/jira/browse/HIVE-18831?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16415746#comment-16415746
]
Rui Li commented on HIVE-18831:
-------------------------------
Hi [~stakiar], thanks for the explanation. I agree differentiating task
exceptions is useful, as described by this JIRA. But I don't feel a strong need
for the new wrapper class. I think it's more natural to expect a Throwable when
something goes wrong. And that's usually what you need at the end of the day,
e.g. when calling {{Task::setException}} and {{JobHandleImpl::setFailure}}. But
returning a Throwable doesn't imply the Throwable has to come from the RPC --
that's implementation details.
A possible improvement similar to SPARK-8625 is try to transfer the original
Throwable and fall back to current implementation when that fails. It can be
done in a separate JIRA though. What do you think?
> Differentiate errors that are thrown by Spark tasks
> ---------------------------------------------------
>
> Key: HIVE-18831
> URL: https://issues.apache.org/jira/browse/HIVE-18831
> Project: Hive
> Issue Type: Sub-task
> Components: Spark
> Reporter: Sahil Takiar
> Assignee: Sahil Takiar
> Priority: Major
> Attachments: HIVE-18831.1.patch, HIVE-18831.2.patch,
> HIVE-18831.3.patch, HIVE-18831.4.patch, HIVE-18831.6.patch, HIVE-18831.7.patch
>
>
> We propagate exceptions from Spark task failures to the client well, but we
> don't differentiate between errors from HS2 / RSC vs. errors thrown by
> individual tasks.
> Main motivation is that when the client sees a propagated Spark exception its
> difficult to know what part of the excution threw the exception.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
