前言
本文介绍了如何使用Shell脚本一键安装Sqoop。Sqoop是一个用于在Apache Hadoop和结构化数据存储(如关系数据库)之间传输数据的工具。通过使用Sqoop,您可以方便地将数据从关系数据库导入到Hadoop集群中进行分析,并将结果数据导出回关系数据库。
在本文中,我们将提供一个Shell脚本,该脚本将自动下载、安装和配置Sqoop。您只需按照步骤执行脚本,即可完成Sqoop的安装和配置过程。
一、使用shell脚本一键安装
1. 复制脚本
首先,将以下脚本内容复制并保存为/tmp/install_sqoop.sh文件。
#!/bin/bash
installDir="/opt/module/sqoop"
sqoop_version="1.4.6"
mysql_connector_version="5.1.37"
if [ -z "$(command -v wget)" ]; then
sudo yum install -y wget
echo "wget安装完成"
fi
if [ -z "$JAVA_HOME" ]; then
echo "JAVA_HOME未设置,请安装jdk1.8,设置Java环境变量再来执行此脚本"
exit 1
fi
if [ ! -d "${installDir}" ]; then
sudo mkdir -p "${installDir}"
if [ $? -eq 0 ]; then
echo "安装目录${installDir}已创建"
else
echo "请确保您有足够的权限来创建目录,请增加权限后再次执行"
exit 1
fi
fi
if [ ! -f /tmp/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha.tar.gz ]; then
wget https://archive.apache.org/dist/sqoop/$sqoop_version/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha.tar.gz -P /tmp/
if [ $? -eq 0 ]; then
echo "sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha.tar.gz下载成功"
else
echo "sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha.tar.gz下载失败,请重试或手动下载到/tmp目录下再次执行"
echo "下载地址:https://archive.apache.org/dist/sqoop/$sqoop_version/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha.tar.gz"
exit 1
fi
fi
if [ -d $installDir/sqoop-$sqoop_version ]; then
echo "$installDir/sqoop-$sqoop_version 已存在,正在删除..."
sudo rm -rf $installDir/sqoop-$sqoop_version
fi
tar -zxvf /tmp/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha.tar.gz -C $installDir
if [ $? -eq 0 ]; then
echo "/tmp/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha.tar.gz解压成功"
else
echo "/tmp/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha.tar.gz解压失败,请查看异常信息后重试"
exit 1
fi
#设置sqoop环境变量
if [ -z "$SQOOP_HOME" ]; then
echo >> ~/.bashrc
echo '#SQOOP_HOME' >> ~/.bashrc
echo "export SQOOP_HOME=$installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha" >> ~/.bashrc
echo 'export PATH=$PATH:$SQOOP_HOME/bin' >> ~/.bashrc
else
echo "SQOOP_HOME已有设置:$SQOOP_HOME"
fi
cp $installDir/sqoop-"$sqoop_version".bin__hadoop-2.0.4-alpha/conf/sqoop-env-template.sh $installDir/sqoop-"$sqoop_version".bin__hadoop-2.0.4-alpha/conf/sqoop-env.sh
if [ $? -ne 0 ]; then
echo "cp $installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/conf/sqoop-env-template.sh $installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/conf/sqoop-env.sh 失败"
exit 1
fi
#设置bin/hadoop所在的路径
if [ -n "$HADOOP_HOME" ]; then
sed -i "s|^#export HADOOP_COMMON_HOME=.*|export HADOOP_COMMON_HOME=$HADOOP_HOME|" "$installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/conf/sqoop-env.sh"
if [ $? -ne 0 ]; then
echo "修改export HADOOP_COMMON_HOME失败"
exit 1
fi
fi
#设置hadoop-*-core.jar可用的路径
if [ -n "$HADOOP_HOME" ]; then
sed -i "s|^#export HADOOP_MAPRED_HOME=.*|export HADOOP_MAPRED_HOME=$HADOOP_HOME|" "$installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/conf/sqoop-env.sh"
if [ $? -ne 0 ]; then
echo "修改export HADOOP_MAPRED_HOME失败"
exit 1
fi
fi
#设置bin/hbase所在路径
if [ -n "$HBASE_HOME" ]; then
sed -i "s|^#export HBASE_HOME=.*|export HBASE_HOME=$HBASE_HOME|" "$installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/conf/sqoop-env.sh"
if [ $? -ne 0 ]; then
echo "修改export HBASE_HOME失败"
exit 1
fi
fi
#设置bin/hive可用的路径
if [ -n "$HIVE_HOME" ]; then
sed -i "s|^#export HIVE_HOME=.*|export HIVE_HOME=$HIVE_HOME|" "$installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/conf/sqoop-env.sh"
if [ $? -ne 0 ]; then
echo "修改export HIVE_HOME失败"
exit 1
fi
fi
#设置zookeeper config dir所在的路径
if [ -n "$ZOOKEEPER_HOME" ]; then
sed -i "s|^#export ZOOCFGDIR=.*|export ZOOCFGDIR=$ZOOKEEPER_HOME/conf|" "$installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/conf/sqoop-env.sh"
if [ $? -ne 0 ]; then
echo "修改export ZOOCFGDIR失败"
exit 1
fi
fi
if [ ! -f /tmp/mysql-connector-java-"$mysql_connector_version".jar ]; then
wget https://repo1.maven.org/maven2/mysql/mysql-connector-java/"$mysql_connector_version"/mysql-connector-java-"$mysql_connector_version".jar -P /tmp
if [ $? -eq 0 ]; then
echo "mysql驱动下载成功"
else
echo "mysql驱动下载失败,请手动下载到/tmp目录下再次执行次脚本"
echo "下载地址:https://repo1.maven.org/maven2/mysql/mysql-connector-java/$mysql_connector_version/mysql-connector-java-$mysql_connector_version.jar"
exit 1
fi
fi
cp /tmp/mysql-connector-java-"$mysql_connector_version".jar $installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/lib
if [ $? -ne 0 ]; then
echo "cp /tmp/mysql-connector-java-"$mysql_connector_version".jar $installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/lib 失败"
exit 1
fi
cp $HIVE_HOME/lib/hive-common-*.jar $installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/lib
if [ $? -ne 0 ]; then
echo "cp $HIVE_HOME/lib/hive-common-*.jar $installDir/sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha/lib 失败"
exit 1
fi
echo "sqoop-$sqoop_version.bin__hadoop-2.0.4-alpha 下载、安装、配置成功"
exit 0
2. 增加执行权限
在终端中执行以下命令,为脚本添加执行权限。
chmod a+x /tmp/install_sqoop.sh
3. 执行脚本
执行以下命令,运行脚本开始下载、安装和配置Kafka。
/tmp/install_sqoop.sh
请等待安装完成,如有异常会有提示。
成功如下图所示:
4. 加载用户环境变量
source ~/.bashrc
5. 查看是否安装成功
sqoop version
成功如下图所示:
总结
通过本文提供的Shell脚本,您可以方便地一键安装Sqoop。Sqoop是一个强大的工具,可用于在Hadoop和关系数据库之间进行数据传输。安装Sqoop后,您可以使用其丰富的功能来导入和导出数据,从而更好地利用Hadoop集群进行数据分析和处理。
希望本教程对您有所帮助!如有任何疑问或问题,请随时在评论区留言。感谢阅读!