Still have 1 requests outstanding when connection from slaveX/X.X.X.X:33202 is closed

ERROR TransportResponseHandler:
Still have 1 requests outstanding when connection from is closed

I added the following parameters

spark.driver.memory=4g
spark.hadoop.mapreduce.input.fileinputformat.split.maxsize=134217728
spark.dynamicAllocation.initialExecutors=50
spark.yarn.executor.memoryOverhead=2048
spark.dynamicAllocation.maxExecutors=1000
spark.dynamicAllocation.tasksPerExecutorSlot=1

The running program is actually logically simple , Just from hive The amount of data read in the table is very large , almost 60+G,
And you need to put some hive Table read to dirver Node , Used to get each executor Mapping values of some data on ,
therefore driver The set resources are large . Exception information thrown at runtime , Check the reason from the Internet. It's probably the server
The number of concurrent connections exceeds its capacity , The server will connect some of these Down fall , That is to say, it is running spark The program , Too many application resources are executed concurrently .

java.util.concurrent.RejectedExecutionException: Task [email protected] rejected from java.util.concurrent.ThreadPoolExecutor
spark Log as shown above . There are other logs ,

18/06/11 18:40:43 ERROR TransportResponseHandler: