[
https://issues.apache.org/jira/browse/SPARK-650?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15578875#comment-15578875
]
Lars Francke commented on SPARK-650:
------------------------------------
I also have to disagree with this being a duplicate or obsolete.
[~oarmand] and [~Skamandros] already mentioned reasons regarding the
duplication.
About it being obsolete: I have seen multiple clients facing this problem,
finding this issue and hoping it'd get fixed some day. I would hesitate a guess
and say that most _users_ of Spark have no JIRA account here and do not
register or log in just to vote for this issue. That said: This issue is (with
six votes) in the top 150 out of almost 17k total issues in the Spark project.
As it happens this is a non-trivial thing to implement in Spark (as far as I
can tell from my limited knowledge of the inner workings) so it's pretty hard
for a "drive by" contributor to help here.
You had the discussion about community perception on the mailing list (re:
Spark Improvement Proposals) and this issue happens to be one of those that at
least I see popping up every once in a while in discussions with clients.
I would love to see this issue staying open as a feature request and have some
hope that someone someday will implement it.
> Add a "setup hook" API for running initialization code on each executor
> -----------------------------------------------------------------------
>
> Key: SPARK-650
> URL: https://issues.apache.org/jira/browse/SPARK-650
> Project: Spark
> Issue Type: New Feature
> Components: Spark Core
> Reporter: Matei Zaharia
> Priority: Minor
>
> Would be useful to configure things like reporting libraries
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
