Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
1.7.0, 1.7.1
Description
When using the StreamingFileSink with S3A backend, occasionally, errors like this will occur:
Caused by: org.apache.flink.fs.s3base.shaded.com.amazonaws.services.s3.model.AmazonS3Exception: Your socket connection to the server was not read from or written to within the timeout period. Idle connections will be closed. (Service: Amazon S3; Status Code: 400; Error Code: RequestTimeout; Request ID: xxx; S3 Extended Request ID: xxx, S3 Extended Request ID: xxx at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutor.handleErrorResponse(AmazonHttpClient.java:1639) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutor.executeOneRequest(AmazonHttpClient.java:1304) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutor.executeHelper(AmazonHttpClient.java:1056)
This causes a restart of flink job, which is often able to recover from, but under heavy load, this can become very frequent.
Turning on debug logs you can find the following relevant stack trace:
2018-12-17 05:55:46,546 DEBUG org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient - FYI: failed to reset content inputstream before throwing up java.io.IOException: Resetting to invalid mark at java.io.BufferedInputStream.reset(BufferedInputStream.java:448) at org.apache.flink.fs.s3base.shaded.com.amazonaws.internal.SdkBufferedInputStream.reset(SdkBufferedInputStream.java:106) at org.apache.flink.fs.s3base.shaded.com.amazonaws.internal.SdkFilterInputStream.reset(SdkFilterInputStream.java:112) at org.apache.flink.fs.s3base.shaded.com.amazonaws.event.ProgressInputStream.reset(ProgressInputStream.java:168) at org.apache.flink.fs.s3base.shaded.com.amazonaws.internal.SdkFilterInputStream.reset(SdkFilterInputStream.java:112) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutor.lastReset(AmazonHttpClient.java:1145) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutor.executeHelper(AmazonHttpClient.java:1070) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutor.doExecute(AmazonHttpClient.java:743) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutor.executeWithTimer(AmazonHttpClient.java:717) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutor.execute(AmazonHttpClient.java:699) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutor.access$500(AmazonHttpClient.java:667) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient$RequestExecutionBuilderImpl.execute(AmazonHttpClient.java:649) at org.apache.flink.fs.s3base.shaded.com.amazonaws.http.AmazonHttpClient.execute(AmazonHttpClient.java:513) at org.apache.flink.fs.s3base.shaded.com.amazonaws.services.s3.AmazonS3Client.invoke(AmazonS3Client.java:4325) at org.apache.flink.fs.s3base.shaded.com.amazonaws.services.s3.AmazonS3Client.invoke(AmazonS3Client.java:4272) at org.apache.flink.fs.s3base.shaded.com.amazonaws.services.s3.AmazonS3Client.doUploadPart(AmazonS3Client.java:3306) at org.apache.flink.fs.s3base.shaded.com.amazonaws.services.s3.AmazonS3Client.uploadPart(AmazonS3Client.java:3291) at org.apache.flink.fs.shaded.hadoop3.org.apache.hadoop.fs.s3a.S3AFileSystem.uploadPart(S3AFileSystem.java:1576) at org.apache.flink.fs.shaded.hadoop3.org.apache.hadoop.fs.s3a.WriteOperationHelper.lambda$uploadPart$8(WriteOperationHelper.java:474) at org.apache.flink.fs.shaded.hadoop3.org.apache.hadoop.fs.s3a.Invoker.once(Invoker.java:109) at org.apache.flink.fs.shaded.hadoop3.org.apache.hadoop.fs.s3a.Invoker.lambda$retry$3(Invoker.java:260) at org.apache.flink.fs.shaded.hadoop3.org.apache.hadoop.fs.s3a.Invoker.retryUntranslated(Invoker.java:317) at org.apache.flink.fs.shaded.hadoop3.org.apache.hadoop.fs.s3a.Invoker.retry(Invoker.java:256) at org.apache.flink.fs.shaded.hadoop3.org.apache.hadoop.fs.s3a.Invoker.retry(Invoker.java:231) at org.apache.flink.fs.shaded.hadoop3.org.apache.hadoop.fs.s3a.WriteOperationHelper.retry(WriteOperationHelper.java:123) at org.apache.flink.fs.shaded.hadoop3.org.apache.hadoop.fs.s3a.WriteOperationHelper.uploadPart(WriteOperationHelper.java:471) at org.apache.flink.fs.s3hadoop.HadoopS3AccessHelper.uploadPart(HadoopS3AccessHelper.java:74) at org.apache.flink.fs.s3.common.writer.RecoverableMultiPartUploadImpl$UploadTask.run(RecoverableMultiPartUploadImpl.java:319) at org.apache.flink.fs.s3.common.utils.BackPressuringExecutor$SemaphoreReleasingRunnable.run(BackPressuringExecutor.java:92) at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149) at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624) at java.lang.Thread.run(Thread.java:748)
This error occurs because of a transient failure in writing a multipart chunk fails and the underlying InputStream cannot be reset. This ResetException should be thrown to the client (as documented here: https://docs.aws.amazon.com/sdk-for-java/v1/developer-guide/best-practices.html) but for some reason is not, instead, the client is retrying the request, but now with a fully consumed InputStream. Because this InputStream is empty/smaller we can't fill up the Content-Length that the multipart upload is expecting, so the socket hangs to eventually be timed out.
This failure happens roughly ~20 times before the AWS client retry logic finally fails the request and the socket time out exception is thrown.
As mentioned in the best practice AWS doc, the best fix for this is to use a File or FileInputStream object or to use the setReadLimit. I tried to use a global SDK property (com.amazonaws.sdk.s3.defaultStreamBufferSize) to set this value, but that did not fix the problem, which I believe is because the InputStream is not mark-able and the AWS client doesn't wrap the stream.
What is confirmed to work is the following patch: https://gist.github.com/addisonj/00fc28f1f8f189380d8e53fdc887fae6
That is obviously not ideal, but it may suffice to just make that configurable.
The other option is to instead expose the S3A WriteHelper option to pass a file to the S3AccessHelper and change the other relevant classes (RefCountedFSOutputStream) to expose the File object and directly hand that to the S3A WriteHelper
Attachments
Issue Links
- links to