Not sure where you get the property name "spark.memory.offHeap.use". The
correct one should be "spark.memory.offHeap.enabled". See
https://spark.apache.org/docs/latest/configuration.html#spark-properties
for details.
On 1/30/23 10:12 AM, Jain, Sanchi wrote:
I am not sure if this is the intended DL for reaching out for help.
Please redirect to the right DL
*From: *Jain, Sanchi <sanchi_j...@comcast.com>
*Date: *Monday, January 30, 2023 at 10:10 AM
*To: *priv...@spark.apache.org <priv...@spark.apache.org>
*Subject: *Request for access to create a jira account- Comcast
Hello there
I am a principal engineer at Comcast and my team is currently working
on building a standalone Spark cluster on a 5 node Linux cluster
environment. We are running into roadblocks due to the following error
observed when a Spark streaming application is submitted to a remote
master.
org.apache.spark.shuffle.MetadataFetchFailedException: Missing an
output location for shuffle 0 partition 11
at
org.apache.spark.MapOutputTracker$.validateStatus(MapOutputTracker.scala:1705)
at
org.apache.spark.MapOutputTracker$.$anonfun$convertMapStatuses$10(MapOutputTracker.scala:1652)
at
org.apache.spark.MapOutputTracker$.$anonfun$convertMapStatuses$10$adapted(MapOutputTracker.scala:1651)
at scala.collection.Iterator.foreach(Iterator.scala:943)
at scala.collection.Iterator.foreach$(Iterator.scala:943)
Here are the other details of the environment configuration –
Software version - spark-3.3.1-bin-hadoop3
Scala version – scala_2.12.15
Total memory assigned to the worker nodes – 14.5 GB (2 GB used)
CPU/Memory assigned to each node – 4 cores/16 GB
Driver memory – 4 G
Executor memory – 3G
Spark-submit command used –
/tmp/spark-3.3.1-bin-hadoop3/bin/spark-submit --master
"spark://<master-host>:7077" --conf spark.submit.deployMode=client
--conf spark.executor.instances=4 --conf spark.executor.memory=3g
--conf spark.driver.memory=4g --conf spark.memory.offHeap.use=true
--conf spark.memory.offHeap.size=3g --conf
spark.sql.broadcastTimeout=300s --conf
spark.sql.autoBroadcastThreshold=1g --class <application-class-name>
./<application-jar-name>.jar
We will really appreciate if we can be assigned a jira account for
submitting an issue in this regard or if we can reach out to the ASF
community for help.
Thanks
Sanchita Jain
sanchita_j...@comcast.com