pipeline.run() blocks when using FlinkRunner

[damccorm]

pipeline.run() is documented to be asynchronous (cf. create-your-pipeline). It seems that when using FlinkRunner (embedded or remote) the call blocks until the pipeline finishes.

Digging into Flink code I found that both, LocalStreamEnvironment and RemoteStreamEnvironment set execution.attached to true. This causes that StreamExecutionEnvironment.execute blocks later on:

    public JobExecutionResult execute(StreamGraph streamGraph) throws Exception {
        final JobClient
jobClient = executeAsync(streamGraph);

        try {
            final JobExecutionResult jobExecutionResult;


           if (configuration.getBoolean(DeploymentOptions.ATTACHED)) {
                jobExecutionResult
= jobClient.getJobExecutionResult().get(); // <==== execution is blocked here
            } else {

               jobExecutionResult = new DetachedJobExecutionResult(jobClient.getJobID());
        
   }

            jobListeners.forEach(
                    jobListener -> jobListener.onJobExecuted(jobExecutionResult,
null));

            return jobExecutionResult;
        } catch (Throwable t) {
            // get()
on the JobExecutionResult Future will throw an ExecutionException. This
            // behaviour was
largely not there in Flink versions before the PipelineExecutor
            // refactoring so we should
strip that exception.
            Throwable strippedException = ExceptionUtils.stripExecutionException(t);


           jobListeners.forEach(
                    jobListener -> {
                        jobListener.onJobExecuted(null,
strippedException);
                    });
            ExceptionUtils.rethrowException(strippedException);


           // never reached, only make javac happy
            return null;
        }
    }

Imported from Jira BEAM-12477. Original Jira may contain additional context.
Reported by: stefan.wachter@gmx.de.

[github-ramA7]

Is there further updates on this. I am looking to use periodic metric pusher. But seems as mentioned in description the client execution is getting blocked. And I see metric pusher class being registered after attaining results object.

[xsolo]

we have the same issues, would like to know if this is on a road map? we use flink 1.15.3

[Abacn]

fixed by #26158