[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17071866#comment-17071866 ] Yu Li commented on FLINK-16018: --- Thanks for the followup and quick fix [~trohrmann]! > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.9.2, 1.10.0 >Reporter: Robert Metzger >Assignee: Till Rohrmann >Priority: Blocker > Labels: pull-request-available > Fix For: 1.9.3, 1.10.1, 1.11.0 > > Time Spent: 20m > Remaining Estimate: 0h > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17070839#comment-17070839 ] Till Rohrmann commented on FLINK-16018: --- As stated before, the underlying problem is that we wait for the {{JobManager}} creation before acknowledging the job submission. Hence, the proper fix would be to make the job submission "non-blocking" (technically is already non blocking but the response is blocked). Since this effort is a bit bigger I would suggest to do the following: * As part of this issue and as a quick fix, we increase {{web.timeout}} to 10 minutes. * I'll create a follow up issue to make the job submission non-blocking which we try to fix asap > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.9.2, 1.10.0 >Reporter: Robert Metzger >Assignee: Till Rohrmann >Priority: Blocker > Fix For: 1.10.1, 1.11.0 > > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17070667#comment-17070667 ] Yu Li commented on FLINK-16018: --- Hi folks, any update on this one? Asking since it's marked as a blocker for 1.10.1 but seems to be stalled. Thanks. > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.9.2, 1.10.0 >Reporter: Robert Metzger >Assignee: Till Rohrmann >Priority: Blocker > Fix For: 1.10.1, 1.11.0 > > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17055566#comment-17055566 ] Zili Chen commented on FLINK-16018: --- A general thought is we respect {{timeout}} parameter in {{Dispatcher#submitJob}}, having a field that helps determine the progress, and complete the future on Timeout with that field(stringified in {{JobSubmissionException}}). > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.9.2, 1.10.0 >Reporter: Robert Metzger >Priority: Blocker > Fix For: 1.10.1, 1.11.0 > > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17054903#comment-17054903 ] Stephan Ewen commented on FLINK-16018: -- Saw another occurrence, raising this to BLOCKER. > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.9.2, 1.10.0 >Reporter: Robert Metzger >Priority: Blocker > Fix For: 1.10.1, 1.11.0 > > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17053579#comment-17053579 ] Zili Chen commented on FLINK-16018: --- [~trohrmann] We internally return immediately on {{JobManagerRunnerFuture}} created instead of the {{JobManagerRunner}} created, and we regard the latter part of job execution. Internally we do it as a follow up of FLINK-10333 because it is included in a whole story we properly handle job status in {{Dispatcher}} and required correctly manipulation of {{JobGraphStore}} & {{JobRegistry}}. > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.11.0 >Reporter: Robert Metzger >Priority: Critical > Fix For: 1.10.1, 1.11.0 > > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17053557#comment-17053557 ] Robert Metzger commented on FLINK-16018: I'm increasing the priority of this ticket to Critical (feel free to disagree with me) Another user opened a ticket (FLINK-16429), and there's a similar report on the user list: https://lists.apache.org/thread.html/r05bd8e13cd7d1b2198bd9e275ff6946af428d4480f75ff5a54200825%40%3Cuser.flink.apache.org%3E > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.11.0 >Reporter: Robert Metzger >Priority: Major > Fix For: 1.10.1, 1.11.0 > > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17039244#comment-17039244 ] Till Rohrmann commented on FLINK-16018: --- A small addition: The creation of the {{JobManager}} is not blocking the main thread of the {{Dispatcher}}. The problem is that we only complete the future returned by {{submitJob}} after the {{JobManager}} has been created and which takes longer than the timeout on the {{RestServer}}. > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.11.0 >Reporter: Robert Metzger >Priority: Major > Fix For: 1.10.1, 1.11.0 > > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17039228#comment-17039228 ] Till Rohrmann commented on FLINK-16018: --- I agree that {{TimeoutExceptions}} should not surface to the user. They are hard to understand and cryptic. I also agree that the root problem is the blocking operation which we should run in a separate thread(pool). Apart from these two aspects, I actually agree with Andrey that we have a problem of un-aligned timeouts here. The {{AskTimeoutException}} the user sees is caused by the {{JobSubmitHandler}} which waits on the {{DisptacherGateway.submitJob}} with a timeout of 10s (default value). The {{JobManager}} creation depends on the Amazon SDK client which, I assume, does some retries and delays in between. Since the 10s are smaller than the overall SDK client timeout, we only see the {{AskTimeoutException}}. This is also underlined by Robert's post where he set {{web.timeout}} to 30s and then sees the actual cause of failure. Retrying the operation in case of a timeout is at the moment not an option since the submit operation is not idempotent. Also, {{DispatcherGateway.submitJob}} currently requires the creation of the {{ExecutionGraph}} before returning. One idea could be to not wait for the creation before acknowledging the job submission. One could define the job submission as that all files have been uploaded to the cluster. This is actually something I wanted to change with FLINK-11719. Another thought could be to say that we require the {{Dispatcher}} to always respond eventually. If this assumption holds, then we could set the {{web.timeout}} to infinity. We do this at other places where we now that the actors run in the same actor system as well. As long as the rest server and the cluster entrypoint run in the same JVM, this assumption might be ok. Of course, the consequence of such a change would be that any kind of (dead/live)locks in downstream components would propagate through to the {{RestServer}}. Maybe setting {{web.timeout}} to 10 minutes would then be a compromise. > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.11.0 >Reporter: Robert Metzger >Priority: Major > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17036957#comment-17036957 ] Stephan Ewen commented on FLINK-16018: -- I don't think this is a misalignment of timeouts. Indeed, when an Akka ask timeout propagates to the user, it means we are not handling failures correctly and simply report back whatever happened in the RPC system. In this specific issue, we also have the problem of blocking/synchronous ExecutionGraph creation. Because some amount of synchronous initialization will most likely always be there, we would need some better way to handle responses to the client. Maybe an additional JobStatus (like INITIALIZING). > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.11.0 >Reporter: Robert Metzger >Priority: Major > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17036944#comment-17036944 ] Robert Metzger commented on FLINK-16018: Thank you for looking at the ticket. I had an offline chat with [~sewen]. He asked me to open this ticket, because there should never be the a case where we show an akka ask timeout to the user. > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.11.0 >Reporter: Robert Metzger >Priority: Major > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17036904#comment-17036904 ] Andrey Zagrebin commented on FLINK-16018: - In this particular case, it was just un-alignment of timeouts which is hard to get between framework and custom operation timeouts. So basically, it can always happen unless submission operation asynchronously monitors some general response timeout for each operation invoked during the submission then it can report where it got stuck. > Improve error reporting when submitting batch job (instead of > AskTimeoutException) > -- > > Key: FLINK-16018 > URL: https://issues.apache.org/jira/browse/FLINK-16018 > Project: Flink > Issue Type: Improvement > Components: Runtime / Coordination >Affects Versions: 1.11.0 >Reporter: Robert Metzger >Priority: Major > > While debugging the {{Shaded Hadoop S3A end-to-end test (minio)}} pre-commit > test, I noticed that the JobSubmission is not producing very helpful error > messages. > Environment: > - A simple batch wordcount job > - a unavailable minio s3 filesystem service > What happens from a user's perspective: > - The job submission fails after 10 seconds with a AskTimeoutException: > {code} > 2020-02-07T11:38:27.1189393Z akka.pattern.AskTimeoutException: Ask timed out > on [Actor[akka://flink/user/dispatcher#-939201095]] after [1 ms]. Message > of type [org.apache.flink.runtime.rpc.messages.LocalFencedMessage]. A typical > reason for `AskTimeoutException` is that the recipient actor didn't send a > reply. > 2020-02-07T11:38:27.1189538Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189616Z at > akka.pattern.PromiseActorRef$$anonfun$2.apply(AskSupport.scala:635) > 2020-02-07T11:38:27.1189713Z at > akka.pattern.PromiseActorRef$$anonfun$1.apply$mcV$sp(AskSupport.scala:648) > 2020-02-07T11:38:27.1189789Z at > akka.actor.Scheduler$$anon$4.run(Scheduler.scala:205) > 2020-02-07T11:38:27.1189883Z at > scala.concurrent.Future$InternalCallbackExecutor$.unbatchedExecute(Future.scala:601) > 2020-02-07T11:38:27.1189973Z at > scala.concurrent.BatchingExecutor$class.execute(BatchingExecutor.scala:109) > 2020-02-07T11:38:27.1190067Z at > scala.concurrent.Future$InternalCallbackExecutor$.execute(Future.scala:599) > 2020-02-07T11:38:27.1190159Z at > akka.actor.LightArrayRevolverScheduler$TaskHolder.executeTask(LightArrayRevolverScheduler.scala:328) > 2020-02-07T11:38:27.1190267Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.executeBucket$1(LightArrayRevolverScheduler.scala:279) > 2020-02-07T11:38:27.1190358Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.nextTick(LightArrayRevolverScheduler.scala:283) > 2020-02-07T11:38:27.1190465Z at > akka.actor.LightArrayRevolverScheduler$$anon$4.run(LightArrayRevolverScheduler.scala:235) > 2020-02-07T11:38:27.1190540Z at java.lang.Thread.run(Thread.java:748) > {code} > What a user would expect: > - An error message indicating why the job submission failed. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (FLINK-16018) Improve error reporting when submitting batch job (instead of AskTimeoutException)
[ https://issues.apache.org/jira/browse/FLINK-16018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17035325#comment-17035325 ] Robert Metzger commented on FLINK-16018: Setting the config to {{web.timeout: 30}} reveals the real underlying issue: {code} 2020-02-07T15:59:57.2209501Z 2020-02-07 15:59:50,547 ERROR org.apache.flink.runtime.dispatcher.StandaloneDispatcher - Failed to submit job 115b0668417af408b4a129499c634396. 2020-02-07T15:59:57.2209618Z java.lang.RuntimeException: org.apache.flink.runtime.client.JobExecutionException: Could not set up JobManager 2020-02-07T15:59:57.2209725Zat org.apache.flink.util.function.CheckedSupplier.lambda$unchecked$0(CheckedSupplier.java:36) 2020-02-07T15:59:57.2209812Zat java.util.concurrent.CompletableFuture$AsyncSupply.run(CompletableFuture.java:1604) 2020-02-07T15:59:57.2209908Zat akka.dispatch.TaskInvocation.run(AbstractDispatcher.scala:40) 2020-02-07T15:59:57.2209993Zat akka.dispatch.ForkJoinExecutorConfigurator$AkkaForkJoinTask.exec(ForkJoinExecutorConfigurator.scala:44) 2020-02-07T15:59:57.2210098Zat akka.dispatch.forkjoin.ForkJoinTask.doExec(ForkJoinTask.java:260) 2020-02-07T15:59:57.2210177Zat akka.dispatch.forkjoin.ForkJoinPool$WorkQueue.runTask(ForkJoinPool.java:1339) 2020-02-07T15:59:57.2210274Zat akka.dispatch.forkjoin.ForkJoinPool.runWorker(ForkJoinPool.java:1979) 2020-02-07T15:59:57.2210433Zat akka.dispatch.forkjoin.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:107) 2020-02-07T15:59:57.2210542Z Caused by: org.apache.flink.runtime.client.JobExecutionException: Could not set up JobManager 2020-02-07T15:59:57.2210631Zat org.apache.flink.runtime.jobmaster.JobManagerRunnerImpl.(JobManagerRunnerImpl.java:152) 2020-02-07T15:59:57.2210744Zat org.apache.flink.runtime.dispatcher.DefaultJobManagerRunnerFactory.createJobManagerRunner(DefaultJobManagerRunnerFactory.java:84) 2020-02-07T15:59:57.2210844Zat org.apache.flink.runtime.dispatcher.Dispatcher.lambda$createJobManagerRunner$6(Dispatcher.java:379) 2020-02-07T15:59:57.2210945Zat org.apache.flink.util.function.CheckedSupplier.lambda$unchecked$0(CheckedSupplier.java:34) 2020-02-07T15:59:57.2211019Z... 7 more 2020-02-07T15:59:57.2211550Z Caused by: org.apache.flink.runtime.client.JobExecutionException: Cannot initialize task 'DataSink (CsvOutputFormat (path: s3://test-data/temp/test_batch_wordcount-0205c494-01da-4cde-ae74-1925833efb57, delimiter: ))': doesBucketExist on test-data: com.amazonaws.SdkClientException: Unable to execute HTTP request: minio: Unable to execute HTTP request: minio 2020-02-07T15:59:57.2211736Zat org.apache.flink.runtime.executiongraph.ExecutionGraphBuilder.buildGraph(ExecutionGraphBuilder.java:216) 2020-02-07T15:59:57.2211831Zat org.apache.flink.runtime.scheduler.SchedulerBase.createExecutionGraph(SchedulerBase.java:253) 2020-02-07T15:59:57.2211942Zat org.apache.flink.runtime.scheduler.SchedulerBase.createAndRestoreExecutionGraph(SchedulerBase.java:225) 2020-02-07T15:59:57.2212028Zat org.apache.flink.runtime.scheduler.SchedulerBase.(SchedulerBase.java:213) 2020-02-07T15:59:57.2212127Zat org.apache.flink.runtime.scheduler.DefaultScheduler.(DefaultScheduler.java:117) 2020-02-07T15:59:57.2212218Zat org.apache.flink.runtime.scheduler.DefaultSchedulerFactory.createInstance(DefaultSchedulerFactory.java:105) 2020-02-07T15:59:57.2212330Zat org.apache.flink.runtime.jobmaster.JobMaster.createScheduler(JobMaster.java:278) 2020-02-07T15:59:57.2212497Zat org.apache.flink.runtime.jobmaster.JobMaster.(JobMaster.java:266) 2020-02-07T15:59:57.2212592Zat org.apache.flink.runtime.jobmaster.factories.DefaultJobMasterServiceFactory.createJobMasterService(DefaultJobMasterServiceFactory.java:98) 2020-02-07T15:59:57.2212714Zat org.apache.flink.runtime.jobmaster.factories.DefaultJobMasterServiceFactory.createJobMasterService(DefaultJobMasterServiceFactory.java:40) 2020-02-07T15:59:57.2212815Zat org.apache.flink.runtime.jobmaster.JobManagerRunnerImpl.(JobManagerRunnerImpl.java:146) 2020-02-07T15:59:57.2212902Z... 10 more 2020-02-07T15:59:57.2213257Z Caused by: org.apache.hadoop.fs.s3a.AWSClientIOException: doesBucketExist on test-data: com.amazonaws.SdkClientException: Unable to execute HTTP request: minio: Unable to execute HTTP request: minio 2020-02-07T15:59:57.2213378Zat org.apache.hadoop.fs.s3a.S3AUtils.translateException(S3AUtils.java:177) 2020-02-07T15:59:57.2213464Zat org.apache.hadoop.fs.s3a.Invoker.once(Invoker.java:111) 2020-02-07T15:59:57.2213565Zat org.apache.hadoop.fs.s3a.Invoker.lambda$retry$3(Invoker.java:260) 2020-02-07T15:59:57.2213641Zat org.apache.hadoop.fs.s3a.Invoker.retryUntranslated(Invoker.java:317) 2020-02-07T15:59:57.2213731Zat org.apache.hadoop.fs.s3a.Invoker.retry(Invoker.java:256) 2020-02-07T15:59:57.2213911Zat org.ap