Spark-class启动脚本解读

#!/usr/bin/env bash

#
# 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.
#

# NOTE: Any changes to this file must be reflected in SparkSubmitDriverBootstrapper.scala!

#判断是否是cygwin环境
cygwin=false
case "`uname`" in
    CYGWIN*) cygwin=true;;
esac

SCALA_VERSION=2.10

# Figure out where Spark is installed
#进去到SPark的安装目录
FWDIR="$(cd `dirname $0`/..; pwd)"

# Export this as SPARK_HOME
# 生成SPARK_HOME环境变量
export SPARK_HOME="$FWDIR"

#执行load-spark-env.sh脚本，主要功能为：
#执行spark-env.sh
#spark-env.sh的主要内容为一些程序过程中的配置和路径的环境变量
. $FWDIR/bin/load-spark-env.sh

#如果没有参数的话执行以下内容
if [ -z "$1" ]; then
  echo "Usage: spark-class <class> [<args>]" 1>&2
  exit 1
fi

#如果SPARK_MEM不为null
if [ -n "$SPARK_MEM" ]; then
  echo -e "Warning: SPARK_MEM is deprecated, please use a more specific config option" 1>&2
  echo -e "(e.g., spark.executor.memory or spark.driver.memory)." 1>&2
fi

# Use SPARK_MEM or 512m as the default memory, to be overridden by specific options
#默认SPARK_MEM的大小为512M
DEFAULT_MEM=${SPARK_MEM:-512m}

SPARK_DAEMON_JAVA_OPTS="$SPARK_DAEMON_JAVA_OPTS -Dspark.akka.logLifecycleEvents=true"

#注意SPARK_DRIVER_MEMORY从spark-env.sh的配置文件中读取SPARK_DRIVER_MEMORY参数

# Add java opts and memory settings for master, worker, history server, executors, and repl.
case "$1" in
  # Master, Worker, and HistoryServer use SPARK_DAEMON_JAVA_OPTS (and specific opts) + SPARK_DAEMON_MEMORY.
  'org.apache.spark.deploy.master.Master')
    OUR_JAVA_OPTS="$SPARK_DAEMON_JAVA_OPTS $SPARK_MASTER_OPTS"
    OUR_JAVA_MEM=${SPARK_DAEMON_MEMORY:-$DEFAULT_MEM}
    ;;
  'org.apache.spark.deploy.worker.Worker')
    OUR_JAVA_OPTS="$SPARK_DAEMON_JAVA_OPTS $SPARK_WORKER_OPTS"
    OUR_JAVA_MEM=${SPARK_DAEMON_MEMORY:-$DEFAULT_MEM}
    ;;
  'org.apache.spark.deploy.history.HistoryServer')
    OUR_JAVA_OPTS="$SPARK_DAEMON_JAVA_OPTS $SPARK_HISTORY_OPTS"
    OUR_JAVA_MEM=${SPARK_DAEMON_MEMORY:-$DEFAULT_MEM}
    ;;

  # Executors use SPARK_JAVA_OPTS + SPARK_EXECUTOR_MEMORY.
  'org.apache.spark.executor.CoarseGrainedExecutorBackend')
    OUR_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_EXECUTOR_OPTS"
    OUR_JAVA_MEM=${SPARK_EXECUTOR_MEMORY:-$DEFAULT_MEM}
    ;;
  'org.apache.spark.executor.MesosExecutorBackend')
    OUR_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_EXECUTOR_OPTS"
    OUR_JAVA_MEM=${SPARK_EXECUTOR_MEMORY:-$DEFAULT_MEM}
    ;;

  # Spark submit uses SPARK_JAVA_OPTS + SPARK_SUBMIT_OPTS +
  # SPARK_DRIVER_MEMORY + SPARK_SUBMIT_DRIVER_MEMORY.
  'org.apache.spark.deploy.SparkSubmit')
    OUR_JAVA_OPTS="$SPARK_JAVA_OPTS $SPARK_SUBMIT_OPTS"
    OUR_JAVA_MEM=${SPARK_DRIVER_MEMORY:-$DEFAULT_MEM}
    if [ -n "$SPARK_SUBMIT_LIBRARY_PATH" ]; then
      OUR_JAVA_OPTS="$OUR_JAVA_OPTS -Djava.library.path=$SPARK_SUBMIT_LIBRARY_PATH"
    fi
    if [ -n "$SPARK_SUBMIT_DRIVER_MEMORY" ]; then
      OUR_JAVA_MEM="$SPARK_SUBMIT_DRIVER_MEMORY"
    fi
    ;;

  *)
    OUR_JAVA_OPTS="$SPARK_JAVA_OPTS"
    OUR_JAVA_MEM=${SPARK_DRIVER_MEMORY:-$DEFAULT_MEM}
    ;;
esac

#找到java的安装目录

# Find the java binary
if [ -n "${JAVA_HOME}" ]; then
  RUNNER="${JAVA_HOME}/bin/java"
else
  if [ `command -v java` ]; then
    RUNNER="java"
  else
    echo "JAVA_HOME is not set" >&2
    exit 1
  fi
fi

# Set JAVA_OPTS to be able to load native libraries and to set heap size
JAVA_OPTS="-XX:MaxPermSize=128m $OUR_JAVA_OPTS"
JAVA_OPTS="$JAVA_OPTS -Xms$OUR_JAVA_MEM -Xmx$OUR_JAVA_MEM"

# Load extra JAVA_OPTS from conf/java-opts, if it exists
if [ -e "$FWDIR/conf/java-opts" ] ; then
  JAVA_OPTS="$JAVA_OPTS `cat $FWDIR/conf/java-opts`"
fi

# Attention: when changing the way the JAVA_OPTS are assembled, the change must be reflected in CommandUtils.scala!

TOOLS_DIR="$FWDIR"/tools

SPARK_TOOLS_JAR=""
if [ -e "$TOOLS_DIR"/target/scala-$SCALA_VERSION/spark-tools*[0-9Tg].jar ]; then
  # Use the JAR from the SBT build
  export SPARK_TOOLS_JAR=`ls "$TOOLS_DIR"/target/scala-$SCALA_VERSION/spark-tools*[0-9Tg].jar`
fi
if [ -e "$TOOLS_DIR"/target/spark-tools*[0-9Tg].jar ]; then
  # Use the JAR from the Maven build
  # TODO: this also needs to become an assembly!
  export SPARK_TOOLS_JAR=`ls "$TOOLS_DIR"/target/spark-tools*[0-9Tg].jar`
fi

# Compute classpath using external script
classpath_output=$($FWDIR/bin/compute-classpath.sh)
if [[ "$?" != "0" ]]; then
  echo "$classpath_output"
  exit 1
else
  CLASSPATH=$classpath_output
fi

if [[ "$1" =~ org.apache.spark.tools.* ]]; then
  if test -z "$SPARK_TOOLS_JAR"; then
    echo "Failed to find Spark Tools Jar in $FWDIR/tools/target/scala-$SCALA_VERSION/" 1>&2
    echo "You need to build spark before running $1." 1>&2
    exit 1
  fi
  CLASSPATH="$CLASSPATH:$SPARK_TOOLS_JAR"
fi

if $cygwin; then
  CLASSPATH=`cygpath -wp $CLASSPATH`
  if [ "$1" == "org.apache.spark.tools.JavaAPICompletenessChecker" ]; then
    export SPARK_TOOLS_JAR=`cygpath -w $SPARK_TOOLS_JAR`
  fi
fi
export CLASSPATH

# In Spark submit client mode, the driver is launched in the same JVM as Spark submit itself.
# Here we must parse the properties file for relevant "spark.driver.*" configs before launching
# the driver JVM itself. Instead of handling this complexity in Bash, we launch a separate JVM
# to prepare the launch environment of this driver JVM.

# 最终调用org.apache.spark.deploy.SparkSubmit类

if [ -n "$SPARK_SUBMIT_BOOTSTRAP_DRIVER" ]; then
  # This is used only if the properties file actually contains these special configs
  # Export the environment variables needed by SparkSubmitDriverBootstrapper
  export RUNNER
  export CLASSPATH
  export JAVA_OPTS
  export OUR_JAVA_MEM
  export SPARK_CLASS=1
  shift # Ignore main class (org.apache.spark.deploy.SparkSubmit) and use our own
  exec "$RUNNER" org.apache.spark.deploy.SparkSubmitDriverBootstrapper "$@"
else
  # Note: The format of this command is closely echoed in SparkSubmitDriverBootstrapper.scala
  if [ -n "$SPARK_PRINT_LAUNCH_COMMAND" ]; then
    echo -n "Spark Command: " 1>&2
    echo "$RUNNER"    #E:Program FilesJavajdk1.7.0_79/bin/java
    echo "$CLASSPATH" #E:cygwin64homehadoop2hivelibmysql-connector-java-5.1.21-bin.jar;E:cygwin64homehadoop2hiveconfhive-site.xml;E:cygwin64homehadoop2spark-1.1.0-bin-hadoop2.4libdatanucleus-core-3.2.2.jar;E:cygwin64homehadoop2spark-1.1.0-bin-hadoop2.4libdatanucleus-api-jdo-3.2.1.jar;E:cygwin64homehadoop2spark-1.1.0-bin-hadoop2.4libdatanucleus-rdbms-3.2.1.jar;.;E:cygwin64usrlocalspark-1.1.0-bin-hadoop2.4conf;E:cygwin64usrlocalspark-1.1.0-bin-hadoop2.4libspark-assembly-1.1.0-hadoop2.4.0.jar;E:cygwin64homehadoop2hadoop-2.5.2etchadoop
    echo $JAVA_OPTS   #-XX:MaxPermSize=512m -Djline.terminal=unix -Xms2048M -Xmx2048M
    echo "$@"         #org.apache.spark.deploy.SparkSubmit --class org.apache.spark.repl.Main spark-shell
    echo "$RUNNER" -cp "$CLASSPATH" $JAVA_OPTS "$@" 1>&2
    echo -e "========================================
" 1>&2
  fi
  exec "$RUNNER" -cp "$CLASSPATH" $JAVA_OPTS "$@"
fi

用Client模式跑一下：

执行一个WordCount: