Hi, I am running spark streaming 1.4.1 on EMR (AMI 3.9) over YARN. The job is reading data from Kinesis and the batch size is of 30s (I used the same value for the kinesis checkpointing). In the executor logs I can see every 5 seconds a sequence of stacktraces indicating that the block replication failed. I am using the default storage level MEMORY_AND_DISK_SER_2. WAL is not enabled nor checkpointing (the checkpoint dir is configured for the spark context but not for the streaming context).
Here is an example of those logs for ip-10-63-160-18. They occur in every executor while trying to replicate to any other executor. 15/10/19 03:11:55 INFO nio.SendingConnection: Initiating connection to [ip-10-63-160-18.ec2.internal/10.63.160.18:50929] 15/10/19 03:11:55 WARN nio.SendingConnection: Error finishing connection to ip-10-63-160-18.ec2.internal/10.63.160.18:50929 java.net.ConnectException: Connection refused at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method) at sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:739) at org.apache.spark.network.nio.SendingConnection.finishConnect(Connection.scala:344) at org.apache.spark.network.nio.ConnectionManager$$anon$10.run(ConnectionManager.scala:292) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) at java.lang.Thread.run(Thread.java:745) 15/10/19 03:11:55 ERROR nio.ConnectionManager: Exception while sending message. java.net.ConnectException: Connection refused at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method) at sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:739) at org.apache.spark.network.nio.SendingConnection.finishConnect(Connection.scala:344) at org.apache.spark.network.nio.ConnectionManager$$anon$10.run(ConnectionManager.scala:292) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) at java.lang.Thread.run(Thread.java:745) 15/10/19 03:11:55 INFO nio.ConnectionManager: Notifying ConnectionManagerId(ip-10-63-160-18.ec2.internal,50929) 15/10/19 03:11:55 INFO nio.ConnectionManager: Handling connection error on connection to ConnectionManagerId(ip-10-63-160-18.ec2.internal,50929) 15/10/19 03:11:55 WARN storage.BlockManager: Failed to replicate input-1-1445242310000 to BlockManagerId(3, ip-10-159-151-22.ec2.internal, 50929), failure #0 java.net.ConnectException: Connection refused at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method) at sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:739) at org.apache.spark.network.nio.SendingConnection.finishConnect(Connection.scala:344) at org.apache.spark.network.nio.ConnectionManager$$anon$10.run(ConnectionManager.scala:292) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) at java.lang.Thread.run(Thread.java:745) 15/10/19 03:11:55 INFO nio.ConnectionManager: Removing SendingConnection to ConnectionManagerId(ip-10-63-160-18.ec2.internal,50929) 15/10/19 03:11:55 INFO nio.SendingConnection: Initiating connection to [ip-10-63-160-18.ec2.internal/10.63.160.18:39506] 15/10/19 03:11:55 WARN nio.SendingConnection: Error finishing connection to ip-10-63-160-18.ec2.internal/10.63.160.18:39506 java.net.ConnectException: Connection refused at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method) at sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:739) at org.apache.spark.network.nio.SendingConnection.finishConnect(Connection.scala:344) at org.apache.spark.network.nio.ConnectionManager$$anon$10.run(ConnectionManager.scala:292) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) at java.lang.Thread.run(Thread.java:745) 15/10/19 03:11:55 ERROR nio.ConnectionManager: Exception while sending message. java.net.ConnectException: Connection refused at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method) at sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:739) at org.apache.spark.network.nio.SendingConnection.finishConnect(Connection.scala:344) at org.apache.spark.network.nio.ConnectionManager$$anon$10.run(ConnectionManager.scala:292) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) at java.lang.Thread.run(Thread.java:745) 15/10/19 03:11:55 INFO nio.ConnectionManager: Notifying ConnectionManagerId(ip-10-63-160-18.ec2.internal,39506) 15/10/19 03:11:55 INFO nio.ConnectionManager: Handling connection error on connection to ConnectionManagerId(ip-10-63-160-18.ec2.internal,39506) 15/10/19 03:11:55 INFO nio.ConnectionManager: Removing SendingConnection to ConnectionManagerId(ip-10-63-160-18.ec2.internal,39506) 15/10/19 03:11:55 WARN storage.BlockManager: Failed to replicate input-1-1445242310000 to BlockManagerId(2, ip-10-141-12-91.ec2.internal, 39506), failure #1 java.net.ConnectException: Connection refused at sun.nio.ch.SocketChannelImpl.checkConnect(Native Method) at sun.nio.ch.SocketChannelImpl.finishConnect(SocketChannelImpl.java:739) at org.apache.spark.network.nio.SendingConnection.finishConnect(Connection.scala:344) at org.apache.spark.network.nio.ConnectionManager$$anon$10.run(ConnectionManager.scala:292) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615) at java.lang.Thread.run(Thread.java:745) 15/10/19 03:11:55 WARN storage.BlockManager: Block input-1-1445242310000 replicated to only 0 peer(s) instead of 1 peers 15/10/19 03:11:55 INFO receiver.BlockGenerator: Pushed block input-1-1445242310000 Thanks, Eugen