Github user mxm commented on a diff in the pull request:
https://github.com/apache/flink/pull/640#discussion_r29431413
--- Diff:
flink-runtime/src/main/scala/org/apache/flink/runtime/jobmanager/JobManager.scala
---
@@ -475,11 +479,21 @@ class JobManager(val flinkConfiguration:
Configuration,
throw new JobSubmissionException(jobId, "The given job is empty")
}
- // see if there already exists an ExecutionGraph for the
corresponding job ID
- executionGraph = currentJobs.getOrElseUpdate(jobGraph.getJobID,
- (new ExecutionGraph(jobGraph.getJobID, jobGraph.getName,
- jobGraph.getJobConfiguration, timeout,
jobGraph.getUserJarBlobKeys, userCodeLoader),
- JobInfo(sender(), System.currentTimeMillis())))._1
+ executionGraph = currentJob match {
+ case Some((graph, _)) if !graph.getState.isTerminalState =>
+ throw new Exception("Job still running")
--- End diff --
I hadn't thought about attaching vertices to a running execution graph.
That could makes sense. Especially, in the case of streaming.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
