[ https://issues.apache.org/jira/browse/SPARK-18052?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15597456#comment-15597456 ]
Sean Owen commented on SPARK-18052: ----------------------------------- This sounds like you have some env or network problem. I'm not aware of any problems of this form and there's no real info on a reproduction here, so I'd generally close this. > Spark Job failing with org.apache.spark.rpc.RpcTimeoutException > --------------------------------------------------------------- > > Key: SPARK-18052 > URL: https://issues.apache.org/jira/browse/SPARK-18052 > Project: Spark > Issue Type: Bug > Components: Spark Submit > Affects Versions: 2.0.0 > Environment: 3 node spark cluster, all AWS r3.xlarge instances > running on ubuntu. > Reporter: Srikanth > Attachments: sparkErrorLog.txt > > > Spark submit jobs are failing with org.apache.spark.rpc.RpcTimeoutException. > increased the spark.executor.heartbeatInterval value from 10s to 60s, but > still the same issue. > This is happening while saving a dataframe to a mounted network drive. Not > using HDFS here. We are able to write successfully for smaller size files > under 10G, the data here we are reading is nearly 20G > driver memory = 10G > executor memory = 25G > Please see the attached log file. -- 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