sqoop 客户端配置

一、配置

  1. 1. 安装 sqoop
  2. sudo apt-get install sqoop
  3. 2. Sqoop 环境
  4. vim ~/.bashrc
  5. # SQOOP
  6. export SQOOP_HOME=/usr/lib/sqoop
  7. export SQOOP_CONF_DIR=$SQOOP_HOME/conf
  8. 3. Sqoop 配置
  9. sudo cp $SQOOP_HOME/conf/sqoop-env-template.sh $SQOOP_HOME/conf/sqoop-env.sh
  10. # 配置 Hive 环境
  11. sudo vim $SQOOP_HOME/conf/sqoop-env.sh
  12. export HIVE_CONF_DIR=$HIVE_HOME/conf
  13. export HADOOP_CLASSPATH=$HADOOP_CLASSPATH:$HIVE_HOME/lib/*
  14. # 软链 hive 配置文件
  15. sudo ln -s $HIVE_HOME/conf/hive-site.xml $SQOOP_HOME/conf/

二、常见问题

  1. 1.出现 org.apache.sqoop.Sqoop 找不到主类
  2. 解决 : sqoop 目录下的 sqoop-1.4.4.jar 拷贝到 hadoop lib 目录下
  3. cd /opt/cloudera/parcels/CDH/lib/hadoop
  4. sudo ln -s ../../jars/sqoop-1.4.5-cdh5.3.3.jar ./
  5. 2.mysql 类加载不到
  6. 解决 : 下载 mysql JDBC 放到 hadoop 目录下即可
  7. cd /opt/cloudera/parcels/CDH/lib/hadoop
  8. sudo ln -s ../../jars/mysql-connector-java-5.1.31.jar ./
  9. 3. HADOOP_MAPRED_HOME is /usr/lib/hadoop-mapreduce 找不到
  10. ERROR tool.ImportTool: Imported Failed: Parameter 'directory' is not a directory
  11. 解决 : sudo ln -s /opt/cloudera/parcels/CDH/lib/hadoop-mapreduce /usr/lib/hadoop-mapreduce
  12. 4. Could not load org.apache.hadoop.hive.conf.HiveConf. Make sure HIVE_CONF_DIR is set correctly
  13. # 配置 Hive 环境
  14. sudo vim $SQOOP_HOME/conf/sqoop-env.sh
  15. export HIVE_CONF_DIR=$HIVE_HOME/conf
  16. export HADOOP_CLASSPATH=$HADOOP_CLASSPATH:$HIVE_HOME/lib/*
  17. # 软链 hive 配置文件
  18. sudo ln -s $HIVE_HOME/conf/hive-site.xml $SQOOP_HOME/conf/