[
https://issues.apache.org/jira/browse/SPARK-15943?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15329581#comment-15329581
]
Sergey Zhemzhitsky commented on SPARK-15943:
--------------------------------------------
Will try to use *spark.memory.useLegacyMode* to see if it helps until spark
1.6.2 is released.
> Spark driver hangs up periodically (cannot receive any reply in 120 seconds)
> ----------------------------------------------------------------------------
>
> Key: SPARK-15943
> URL: https://issues.apache.org/jira/browse/SPARK-15943
> Project: Spark
> Issue Type: Bug
> Components: Spark Core
> Affects Versions: 1.6.0
> Environment: Ubuntu Server 14.04
> Spark 1.6.0
> Hadoop 2.6.0
> Reporter: Sergey Zhemzhitsky
> Priority: Critical
> Labels: spark
> Attachments: driver-stacktrace.log, executor-threaddump-jstack.log,
> executor-threaddump.log
>
>
> Spark driver hangs up periodically with the following stacktrace on driver
> [^driver-stacktrace.log] and threaddump on the executor that causes this
> issue [^executor-threaddump.log]
> The attached [^executor-threaddump.log] has been retrieved by means of spark
> user interface. Trying to dump the threads by means of *jstack -l -F <pid>*
> allowed us to retrieve only the following stacktrace
> [^executor-threaddump-jstack.log]
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
