[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: JobManager did not respond within 60000 ms


You have quite complicated job graph and very low memory settings for the job manager and task manager. It might be that long GC pauses are causing this problem.

Secondly, there are quite some results in google search of this error that points toward high-availability issues. Have you read those previously reported problems? 

Thanks, Piotrek 

On 9 Oct 2018, at 09:57, jpreisner@xxxxxxx wrote:

I have a streaming job that works in standalone cluster. Flink version is 1.4.1. Everything was working so far. But since I added new treatments, I can not start my job anymore. I have this exception :

org.apache.flink.client.program.ProgramInvocationException: The program execution failed: JobManager did not respond within 60000 ms
       at org.apache.flink.client.program.ClusterClient.runDetached(
       at org.apache.flink.client.program.StandaloneClusterClient.submitJob(
       at org.apache.flink.client.program.DetachedEnvironment.finalizeExecute(
       at org.apache.flink.client.CliFrontend.executeProgram(
       at org.apache.flink.client.CliFrontend.parseParameters(
       at org.apache.flink.client.CliFrontend$
       at org.apache.flink.client.CliFrontend$
       at org.apache.flink.client.CliFrontend.main(
Caused by: org.apache.flink.runtime.client.JobTimeoutException: JobManager did not respond within 60000 ms
       at org.apache.flink.runtime.client.JobClient.submitJobDetached(
       at org.apache.flink.client.program.ClusterClient.runDetached(
       ... 11 more
Caused by: java.util.concurrent.TimeoutException
       at java.util.concurrent.CompletableFuture.timedGet(
       at java.util.concurrent.CompletableFuture.get(
       at org.apache.flink.runtime.client.JobClient.submitJobDetached(
       ... 12 more

I see a very strange behavior. When I comment on a function (any one, for example a FilterFunction, which was present before or after my modification).
I tried to change the configuration (akka.client.timeout and akka.framesize) without success.

This is my flink-conf.yaml
 jobmanager.rpc.address: myhost
 jobmanager.rpc.port: 6123
 jobmanager.heap.mb: 128
 taskmanager.heap.mb: 1024
 taskmanager.numberOfTaskSlots: 100
 taskmanager.memory.preallocate: false 6121
 parallelism.default: 1
 taskmanager.tmp.dirs: /dohdev/flink/tmp/tskmgr /dohdev/flink/tmp/blob
 jobmanager.web.port: -1
 high-availability: zookeeper
 high-availability.zookeeper.quorum: localhost:2181
 high-availability.zookeeper.path.root: /dohdev/flink
 high-availability.cluster-id: dev
 high-availability.storageDir: file:////mnt/metaflink
 high-availability.zookeeper.storageDir: /mnt/metaflink/inh/agregateur/recovery
 restart-strategy: fixed-delay
 restart-strategy.fixed-delay.attempts: 1000
 restart-strategy.fixed-delay.delay: 5 s
 zookeeper.sasl.disable: true
 blob.service.cleanup.interval: 60

And I launch a job with this command : bin/flink run -d myjar.jar

I added as an attachment a graph of my job when it works (Graph.PNG).

Do you have an idea of the problem ?