[FLINK-5355] Handle AmazonKinesisException gracefully in Kinesis Streaming Connector

• Type: Improvement
• Status: Resolved
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 1.1.3, 1.2.0
• Component/s: Connectors / Kinesis
• Assignee: Scott Kidder
• Reporter: Scott Kidder
• Created: 16/Dec/16 15:21
• Updated: 14/Feb/19 13:27
• Resolved: 20/Jan/17 15:56

Description

My Flink job that consumes from a Kinesis stream must be restarted at least once daily due to an uncaught AmazonKinesisException when reading from Kinesis. The complete stacktrace looks like:

com.amazonaws.services.kinesis.model.AmazonKinesisException: null (Service: AmazonKinesis; Status Code: 500; Error Code: InternalFailure; Request ID: dc1b7a1a-1b97-1a32-8cd5-79a896a55223)
	at com.amazonaws.http.AmazonHttpClient$RequestExecutor.handleErrorResponse(AmazonHttpClient.java:1545)
	at com.amazonaws.http.AmazonHttpClient$RequestExecutor.executeOneRequest(AmazonHttpClient.java:1183)
	at com.amazonaws.http.AmazonHttpClient$RequestExecutor.executeHelper(AmazonHttpClient.java:964)
	at com.amazonaws.http.AmazonHttpClient$RequestExecutor.doExecute(AmazonHttpClient.java:676)
	at com.amazonaws.http.AmazonHttpClient$RequestExecutor.executeWithTimer(AmazonHttpClient.java:650)
	at com.amazonaws.http.AmazonHttpClient$RequestExecutor.execute(AmazonHttpClient.java:633)
	at com.amazonaws.http.AmazonHttpClient$RequestExecutor.access$300(AmazonHttpClient.java:601)
	at com.amazonaws.http.AmazonHttpClient$RequestExecutionBuilderImpl.execute(AmazonHttpClient.java:583)
	at com.amazonaws.http.AmazonHttpClient.execute(AmazonHttpClient.java:447)
	at com.amazonaws.services.kinesis.AmazonKinesisClient.doInvoke(AmazonKinesisClient.java:1747)
	at com.amazonaws.services.kinesis.AmazonKinesisClient.invoke(AmazonKinesisClient.java:1723)
	at com.amazonaws.services.kinesis.AmazonKinesisClient.getRecords(AmazonKinesisClient.java:858)
	at org.apache.flink.streaming.connectors.kinesis.proxy.KinesisProxy.getRecords(KinesisProxy.java:193)
	at org.apache.flink.streaming.connectors.kinesis.internals.ShardConsumer.getRecords(ShardConsumer.java:268)
	at org.apache.flink.streaming.connectors.kinesis.internals.ShardConsumer.run(ShardConsumer.java:176)
	at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
	at java.lang.Thread.run(Thread.java:745)

It's interesting that the Kinesis endpoint returned a 500 status code, but that's outside the scope of this issue.

I think we can handle this exception in the same manner as a ProvisionedThroughputException: performing an exponential backoff and retrying a finite number of times before throwing an exception.

Issue Links

• links to: GitHub Pull Request #3078