Thankyou @venu123 wrote: Thank yuo, i will second option @schuberth wrote: Hello Venu,   The spill messages and the log snippet indicate that the Hive's MapReduce task is using disk to sort data because the buffer allocated for sorting is full. There's couple of things that you can tune:   1. Increase the container memory allocated to Map tasks (remember to increase the heap size of the map task too!) 2. Increase the sort buffer size (mapreduce.task.io.sort.mb)   Hope this helps.   i will try second option ... View more

Try adding shell script in distributed cache   <workflow-app name="Locus_Ingest_to_Staging_to_HDFS_-_Shell" xmlns="uri:oozie:workflow:0.5"> <start to="shell-5cb7"/> <kill name="Kill"> <message>Action failed, error message[${wf:errorMessage(wf:lastErrorNode())}]</message> </kill> <action name="shell-5cb7"> <shell xmlns="uri:oozie:shell-action:0.1"> <job-tracker>${jobTracker}</job-tracker> <name-node>${nameNode}</name-node> <exec>ssh_local_to_hdfs.sh</exec> <argument>&quot;/opt/ingest/locus/&quot;</argument> <argument>&quot;/data/ingest/locus/&quot;</argument> <argument>&quot;vax;1&quot;</argument> <argument>&quot;vaxtoken&quot;</argument> <file>hdfspath/ssh_local_to_hdfs.sh</file> <capture-output/> </shell> <ok to="End"/> <error to="Kill"/> </action> <end name="End"/> </workflow-app>   ... View more

But this error will not cause a role crash as your title describes. It likely went down either cause framed transport is not enabled and a bad request reached the thrift server, or maybe an OOME (see stdout of the crashed process). ... View more