You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Ching Lin (Jira)" <ji...@apache.org> on 2019/10/22 09:56:00 UTC
[jira] [Commented] (SPARK-29548) Redirect system print stream to
log4j and improve robustness
[ https://issues.apache.org/jira/browse/SPARK-29548?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16956885#comment-16956885 ]
Ching Lin commented on SPARK-29548:
-----------------------------------
how about using checkpoint instead of log4j ?
> Redirect system print stream to log4j and improve robustness
> ------------------------------------------------------------
>
> Key: SPARK-29548
> URL: https://issues.apache.org/jira/browse/SPARK-29548
> Project: Spark
> Issue Type: Improvement
> Components: Spark Core
> Affects Versions: 3.0.0
> Reporter: jiaan.geng
> Priority: Major
>
> In a production environment, user behavior is highly random and uncertain.
> For example: Users use `System.out` or `System.err` to print information.
> But the system print stream may cause some trouble, such as: the disk file is too large.
> In my production environment, it causes the disk to be full and let [NodeManager] works not fine.
> A method of threat is to forbid the use of `System.out` or `System.err`. But unfriendly to the users.
> A better method is to redirecting the system print stream to `Log4j` and Spark can take advantage of `Log4j`'s split log.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org