You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Apache Spark (Jira)" <ji...@apache.org> on 2022/06/30 00:47:00 UTC
[jira] [Commented] (SPARK-39636) Fix multiple small bugs in JsonProtocol, impacting StorageLevel and Task/Executor resource requests
[ https://issues.apache.org/jira/browse/SPARK-39636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17560689#comment-17560689 ]
Apache Spark commented on SPARK-39636:
--------------------------------------
User 'JoshRosen' has created a pull request for this issue:
https://github.com/apache/spark/pull/37027
> Fix multiple small bugs in JsonProtocol, impacting StorageLevel and Task/Executor resource requests
> ---------------------------------------------------------------------------------------------------
>
> Key: SPARK-39636
> URL: https://issues.apache.org/jira/browse/SPARK-39636
> Project: Spark
> Issue Type: Improvement
> Components: Spark Core, Web UI
> Affects Versions: 1.0.0
> Reporter: Josh Rosen
> Assignee: Josh Rosen
> Priority: Major
>
> While working on SPARK-39489, I spotted three pre-existing bugs in JsonProtocol:
> * {{TaskResourceRequest}} loses precision for values < 0.5. The {{amount}} is a floating point number which is either between 0 and 0.5 or is a positive integer, but the read path assumes it is an integer.
> * {{ExecutorResourceRequest}} integer overflows for values larger than an integer because the write path writes longs but the read path assumes integers.
> * The {{OFF_HEAP}} storage level is not handled properly: the {{offHeap}} field isn't included in the JSON, so this StorageLevel cannot be round-tripped through JSON. This could cause the History Server to display inaccurate "off heap memory used" stats on the executors page.
> I'm submitting a separate PR to fix those issues.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org