tillrohrmann commented on a change in pull request #18220: URL: https://github.com/apache/flink/pull/18220#discussion_r775870364
########## File path: flink-clients/src/main/java/org/apache/flink/client/program/AsyncClusterClientProvider.java ########## @@ -0,0 +1,44 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + * + */ + +package org.apache.flink.client.program; + +import org.apache.flink.annotation.Internal; + +/** Provider for {@link ClusterClient ClusterClients}. */ +@Internal +public abstract class AsyncClusterClientProvider<T> implements ClusterClientProvider<T> { Review comment: I think it would be better to introduce an `AsyncClusterClientProvider` interface that offers a `getClusterClientAsync` method that returns a `CompletableFuture<ClusterClient<T>>`. It can extend the `ClusterClientProvider`. The default implementation of `getClusterClient` could then wait on this future. ########## File path: flink-yarn/src/main/java/org/apache/flink/yarn/YarnClusterDescriptor.java ########## @@ -1195,9 +1220,21 @@ private ApplicationReport startAppMaster( yarnClient.submitApplication(appContext); LOG.info("Waiting for the cluster to be allocated"); + + ApplicationReport report = waitForAppSubmitted(appId, false); + + // since deployment was successful, remove the hook + ShutdownHookUtil.removeShutdownHook(deploymentFailureHook, getClass().getSimpleName(), LOG); + return report; + } + + private ApplicationReport waitForAppSubmitted( Review comment: Maybe it is better to create an `ApplicationReportProvider` that encapsulates the logic for how to retrieve the application report and until when to wait. Then the caller of `startAppMaster` can decide whether it wants to wait on it or not. ########## File path: flink-yarn/src/main/java/org/apache/flink/yarn/YarnClusterDescriptor.java ########## @@ -613,21 +617,42 @@ public void killCluster(ApplicationId applicationId) throws FlinkException { yarnApplication, validClusterSpecification); - // print the application id for user to cancel themselves. - if (detached) { - final ApplicationId yarnApplicationId = report.getApplicationId(); - logDetachedClusterInformation(yarnApplicationId, LOG); - } + final ApplicationId yarnApplicationId = report.getApplicationId(); + applicationId = yarnApplicationId; + + AsyncClusterClientProvider<ApplicationId> clusterClientProvider = Review comment: I don't think that we should create annonymous inner classes. This gives access to implementation details we should not expose. E.g. what if the `flinkConfiguration` changes? I think it would be better to properly separate the `AsyncClusterClientProvider` implementation from the `YarnClusterDescriptor`. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: issues-unsubscr...@flink.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
