Dolphin scheduler 2.0.5 spark task test summary (source code optimization)

spark Task test summary

---

Various problems encountered during the test , See details I for deployment and definition and Detail II , The manual execution on the environment succeeded , Theoretically, there is no problem with scheduling , Then copy the task scheduling command to the environment and execute it successfully , But the task scheduling itself fails . Finally, it is found that the parameter is too long, resulting in newline , Thus the execution fails .

Modify source code （SparkArgsUtils）

---

• When the parameter is too long , It's easy to enter a carriage return line feed , To prevent input from entering a carriage return , The background shall be replaced uniformly
  

• SparkArgsUtils

/* * 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.dolphinscheduler.plugin.task.spark;

import org.apache.dolphinscheduler.plugin.task.util.ArgsUtils;
import org.apache.dolphinscheduler.spi.task.ResourceInfo;
import org.apache.dolphinscheduler.spi.utils.StringUtils;

import java.util.ArrayList;
import java.util.List;

/** * spark args utils */
public class SparkArgsUtils {
    

    private static final String SPARK_CLUSTER = "cluster";

    private static final String SPARK_LOCAL = "local";

    private static final String SPARK_ON_YARN = "yarn";

    private SparkArgsUtils() {
    
        throw new IllegalStateException("Utility class");
    }

    /** * build args * * @param param param * @return argument list */
    public static List<String> buildArgs(SparkParameters param) {
    
        List<String> args = new ArrayList<>();
        args.add(SparkConstants.MASTER);

        String deployMode = StringUtils.isNotEmpty(param.getDeployMode()) ? param.getDeployMode() : SPARK_CLUSTER;
        if (!SPARK_LOCAL.equals(deployMode)) {
    
            args.add(SPARK_ON_YARN);
            args.add(SparkConstants.DEPLOY_MODE);
        }
        args.add(deployMode);

        ProgramType programType = param.getProgramType();
        String mainClass = param.getMainClass();
        if (programType != null && programType != ProgramType.PYTHON && StringUtils.isNotEmpty(mainClass)) {
    
            args.add(SparkConstants.MAIN_CLASS);
            args.add(mainClass);
        }

        int driverCores = param.getDriverCores();
        if (driverCores > 0) {
    
            args.add(SparkConstants.DRIVER_CORES);
            args.add(String.format("%d", driverCores));
        }

        String driverMemory = param.getDriverMemory();
        if (StringUtils.isNotEmpty(driverMemory)) {
    
            args.add(SparkConstants.DRIVER_MEMORY);
            args.add(driverMemory);
        }

        int numExecutors = param.getNumExecutors();
        if (numExecutors > 0) {
    
            args.add(SparkConstants.NUM_EXECUTORS);
            args.add(String.format("%d", numExecutors));
        }

        int executorCores = param.getExecutorCores();
        if (executorCores > 0) {
    
            args.add(SparkConstants.EXECUTOR_CORES);
            args.add(String.format("%d", executorCores));
        }

        String executorMemory = param.getExecutorMemory();
        if (StringUtils.isNotEmpty(executorMemory)) {
    
            args.add(SparkConstants.EXECUTOR_MEMORY);
            args.add(executorMemory);
        }

        String appName = param.getAppName();
        if (StringUtils.isNotEmpty(appName)) {
    
            args.add(SparkConstants.SPARK_NAME);
            args.add(ArgsUtils.escape(appName));
        }

        String others = param.getOthers();
        if (!SPARK_LOCAL.equals(deployMode) && (StringUtils.isEmpty(others) || !others.contains(SparkConstants.SPARK_QUEUE))) {
    
            String queue = param.getQueue();
            if (StringUtils.isNotEmpty(queue)) {
    
                args.add(SparkConstants.SPARK_QUEUE);
                args.add(queue);
            }
        }

        // --conf --files --jars --packages
        if (StringUtils.isNotEmpty(others)) {
    
            args.add(others);
        }

        ResourceInfo mainJar = param.getMainJar();
        if (mainJar != null) {
    
            args.add(mainJar.getRes());
        }

        String mainArgs = param.getMainArgs();
        if (StringUtils.isNotEmpty(mainArgs)) {
    
            // renxiaozhao 20220615  Prevent command line breaks 
            args.add(mainArgs.replace("\r", " ").replace("\n", " "));
        }

        return args;
    }

}

• Post replacement log （ Entering carriage return will not break the line ）
  

Definition of environment variables

---

The resource center has uploaded the file , You can only see the file name , Can't see hdfs The full path , All path variables can be defined uniformly , For users, you can splice file names directly through full path variables

• Define global variables
  
• Reference global variables
  
• journal ( Result output parameter splicing date , Prevent reporting that there is an error in the directory )