Github user pwendell commented on a diff in the pull request:
https://github.com/apache/spark/pull/86#discussion_r10506918
--- Diff: core/src/main/scala/org/apache/spark/deploy/SparkSubmit.scala ---
@@ -0,0 +1,188 @@
+/*
+ * 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.spark.deploy
+
+import java.io.File
+import java.net.URL
+import java.net.URLClassLoader
+
+import scala.collection.mutable.ArrayBuffer
+import scala.collection.mutable.HashMap
+import scala.collection.mutable.Map
+
+object SparkSubmit {
+ val YARN = 1
+ val STANDALONE = 2
+ val MESOS = 4
+ val LOCAL = 8
+ val ALL_CLUSTER_MGRS = YARN | STANDALONE | MESOS | LOCAL
+
+ var clusterManager: Int = LOCAL
+
+ def main(args: Array[String]) {
+ val appArgs = new SparkSubmitArguments(args)
+ val (childArgs, classpath, sysProps, mainClass) =
createLaunchEnv(appArgs)
+ launch(childArgs, classpath, sysProps, mainClass)
+ }
+
+ /**
+ * @return
+ * a tuple containing the arguments for the child, a list of
classpath
+ * entries for the child, and the main class for the child
+ */
+ def createLaunchEnv(appArgs: SparkSubmitArguments): (ArrayBuffer[String],
+ ArrayBuffer[String], Map[String, String], String) = {
+ if (appArgs.master != null) {
+ if (appArgs.master.startsWith("yarn")) {
+ clusterManager = YARN
+ } else if (appArgs.master.startsWith("spark")) {
+ clusterManager = STANDALONE
+ } else if (appArgs.master.startsWith("mesos")) {
+ clusterManager = MESOS
+ } else if (appArgs.master.startsWith("local")) {
+ clusterManager = LOCAL
+ } else {
+ System.err.println("master must start with yarn, mesos, spark, or
local")
+ System.exit(1)
+ }
+ }
+
+ val deployOnCluster = appArgs.deployMode == "cluster"
+ val childClasspath = new ArrayBuffer[String]()
+ val childArgs = new ArrayBuffer[String]()
+ val sysProps = new HashMap[String, String]()
+ var childMainClass = ""
+
+ if (clusterManager == MESOS && deployOnCluster) {
+ System.err.println("Mesos does not support running the driver on the
cluster")
+ System.exit(1)
+ }
+
+ if (!deployOnCluster) {
+ childMainClass = appArgs.mainClass
+ childClasspath += appArgs.primaryResource
+ } else if (clusterManager == YARN) {
+ childMainClass = "org.apache.spark.deploy.yarn.Client"
+ childArgs += ("--jar", appArgs.primaryResource)
+ childArgs += ("--class", appArgs.mainClass)
+ }
+
+ val options = List[OptionAssigner](
+ new OptionAssigner(appArgs.driverMemory, YARN, true, clOption =
"--master-memory"),
+ new OptionAssigner(appArgs.name, YARN, true, clOption = "--name"),
+ new OptionAssigner(appArgs.queue, YARN, true, clOption = "--queue"),
+ new OptionAssigner(appArgs.queue, YARN, false, sysProp =
"spark.yarn.queue"),
+ new OptionAssigner(appArgs.numExecutors, YARN, true, clOption =
"--num-workers"),
+ new OptionAssigner(appArgs.numExecutors, YARN, false, sysProp =
"spark.worker.instances"),
+ new OptionAssigner(appArgs.executorMemory, YARN, true, clOption =
"--worker-memory"),
+ new OptionAssigner(appArgs.executorMemory, STANDALONE, true,
clOption = "--memory"),
+ new OptionAssigner(appArgs.executorMemory, STANDALONE | MESOS |
YARN, false, sysProp = "spark.executor.memory"),
+ new OptionAssigner(appArgs.executorCores, YARN, true, clOption =
"--worker-cores"),
+ new OptionAssigner(appArgs.executorCores, STANDALONE, true, clOption
= "--cores"),
+ new OptionAssigner(appArgs.executorCores, STANDALONE | MESOS | YARN,
false, sysProp = "spark.cores.max"),
+ new OptionAssigner(appArgs.files, YARN, false, sysProp =
"spark.yarn.dist.files"),
+ new OptionAssigner(appArgs.files, YARN, true, clOption = "--files"),
+ new OptionAssigner(appArgs.archives, YARN, false, sysProp =
"spark.yarn.dist.archives"),
+ new OptionAssigner(appArgs.archives, YARN, true, clOption =
"--archives"),
+ new OptionAssigner(appArgs.moreJars, YARN, true, clOption =
"--addJars")
+ )
+
+ // more jars
+ if (appArgs.moreJars != null && !deployOnCluster) {
+ for (jar <- appArgs.moreJars.split(",")) {
+ childClasspath += jar
+ }
+ }
+
+ for (opt <- options) {
+ if (opt.value != null && deployOnCluster == opt.deployOnCluster &&
+ (clusterManager & opt.clusterManager) != 0) {
+ if (opt.clOption != null) {
+ childArgs += (opt.clOption, opt.value)
+ } else if (opt.sysProp != null) {
+ sysProps.put(opt.sysProp, opt.value)
+ }
+ }
+ }
+
+ // supervise
+ if (appArgs.supervise && deployOnCluster && clusterManager ==
STANDALONE) {
+ childArgs += "--supervise"
+ }
+
+ if (deployOnCluster && clusterManager == STANDALONE) {
--- End diff --
Would it make sense to nest the supervise thing inside of this if statement:
if (appArgs.supervise) childArgs += "--supervise"
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
