flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From GitBox <...@apache.org>
Subject [GitHub] [flink] zentol commented on a change in pull request #8498: [FLINK-12413] [runtime] Implement ExecutionFailureHandler
Date Thu, 23 May 2019 11:28:32 GMT
zentol commented on a change in pull request #8498: [FLINK-12413] [runtime] Implement ExecutionFailureHandler
URL: https://github.com/apache/flink/pull/8498#discussion_r286893662
 
 

 ##########
 File path: flink-runtime/src/main/java/org/apache/flink/runtime/executiongraph/failover/flip1/ExecutionFailureHandler.java
 ##########
 @@ -0,0 +1,89 @@
+/*
+ * 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.runtime.executiongraph.failover.flip1;
+
+import org.apache.flink.annotation.VisibleForTesting;
+import org.apache.flink.runtime.JobException;
+import org.apache.flink.runtime.scheduler.strategy.ExecutionVertexID;
+import org.apache.flink.runtime.throwable.ThrowableClassifier;
+import org.apache.flink.runtime.throwable.ThrowableType;
+
+import static org.apache.flink.util.Preconditions.checkNotNull;
+
+/**
+ * This handler deals with task failures to return a {@link FailureHandlingResult} which
contains tasks
+ * to restart to recover from failures.
+ */
+public class ExecutionFailureHandler {
+
+	/** Strategy to judge which tasks should be restarted. */
+	private FailoverStrategy failoverStrategy;
+
+	/** Strategy to judge whether and when a restarting should be done. */
+	private RestartBackoffTimeStrategy restartBackoffTimeStrategy;
+
+	/**
+	 * Creates the handler to deal with task failures.
+	 *
+	 * @param topology of tasks of current job
+	 * @param failoverStrategyFactory is the factory to create the failover strategy
+	 * @param restartStrategyFactory is the factory to create the restart strategy
+	 */
+	public ExecutionFailureHandler(
+		FailoverTopology topology,
+		FailoverStrategy.Factory failoverStrategyFactory,
+		RestartBackoffTimeStrategy.Factory restartStrategyFactory) {
+
+		this.failoverStrategy = checkNotNull(failoverStrategyFactory.create(topology));
+		this.restartBackoffTimeStrategy = checkNotNull(restartStrategyFactory.create());
+	}
+
+	/**
+	 * Return result of failure handling. Can be a set of task vertices to restart
+	 * and a delay of the restarting. Or that the failure is not recoverable and the reason
for it.
+	 *
+	 * @param failedTask is the ID of the failed task vertex
+	 * @param cause of the task failure
+	 * @return result of the failure handling
+	 */
+	public FailureHandlingResult getFailureHandlingResult(ExecutionVertexID failedTask, Throwable
cause) {
+		if (ThrowableClassifier.getThrowableType(cause) == ThrowableType.NonRecoverableError) {
+			return new FailureHandlingResult(new JobException("The failure is not recoverable", cause));
+		}
+
+		restartBackoffTimeStrategy.notifyFailure(cause);
+		if (restartBackoffTimeStrategy.canRestart()) {
+			return new FailureHandlingResult(
+				failoverStrategy.getTasksNeedingRestart(failedTask, cause),
+				restartBackoffTimeStrategy.getBackoffTime());
+		} else {
+			return new FailureHandlingResult(
+				new JobException("Failed task restarting is suppressed by " + restartBackoffTimeStrategy,
cause));
+		}
+	}
+
+	@VisibleForTesting
+	FailoverStrategy getFailoverStrategy() {
 
 Review comment:
   this method (and `getRestartBackoffTimeStrategy`) seem unnecessary; any test using this
class must pass factories for the strategies, and thus can already access the actual strategy
instance.
   This could be as simple as:
   ```
   FailoverStrategy failoverStrategy = new MyFailoverStrategy(failoverTopology);
   RestartBackoffTimeStrategy restartStrategy = new MyRestartBackoffTimeStrategy();
   
   ExecutionFailureHandler handler = new ExecutionFailureHandler(
   	failoverTopology,
   	topology -> failoverStrategy,
   	() -> restartStrategy);
   ```

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services

Mime
View raw message