[ https://issues.apache.org/jira/browse/FLINK-17788?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17116780#comment-17116780 ]
Kostas Kloudas commented on FLINK-17788: ---------------------------------------- [~zjffdu] can you be more specific on what you see as error? Also do you launch the Yarn cluster from the Flink Shell? Or you expect to find a running Yarn Cluster? And if the second is the case, how do you specify the {{ApplicationId}}? The reason I am asking is because the {{DeploymentOptions.TARGET}} that you remove in the PR has to be set to something. If you launch the yarn cluster from the ScalaShell (the {{FlinkShell.deployNewYarnCluster()}} method), then I assume that Flink will find the "properties" file that contains the ApplicationId of the cluster and it will know where to submit the job. > scala shell in yarn mode is broken > ---------------------------------- > > Key: FLINK-17788 > URL: https://issues.apache.org/jira/browse/FLINK-17788 > Project: Flink > Issue Type: Bug > Components: Scala Shell > Affects Versions: 1.11.0 > Reporter: Jeff Zhang > Priority: Major > Labels: pull-request-available > > When I start scala shell in yarn mode, one yarn app will be launched, and > after I write some flink code and trigger a flink job, another yarn app will > be launched but would failed to launch due to some conflicts. -- This message was sent by Atlassian Jira (v8.3.4#803005)