Hadoop完整伪分布式搭建(包含zookeeper+hbase+hive)

1.jdk
2.jdk环境变量
3.hadoop
4.hadoop环境变量
5.hadoop文件修改
a.core-site.xml
1. Fs.defaultFS = hdfs://192.168.220.129:9000
2. hadoop.tmp.dir=/usr/local/softwave/hadoop-2.7.3/tmp
b.hdfs-site.xml
1. Dfs.replication = 1
c.mapred-site.xml
1. Mapreduce.framewok.name=yarn
d.yarn-site.xml
a) Yarn.resourcemanager.localhost=localhost
b) Yarn.nodemanager.aux-service=mapreduce_shuffle
e.hadoop-env.sh
JAVA_HOME
6.hdfs namenode -format
7.hive 环境变量
hive-site


javax.jdo.option.ConnectionURL
jdbc:mysql://192.168.1.5:3306/hive


javax.jdo.option.ConnectionDriverName
com.mysql.jdbc.Driver


javax.jdo.option.ConnectionUserName
root


javax.jdo.option.ConnectionPassword
ok


初始化hive, 将mysql作为hive的元数据库
!schematool -dbType mysql -initSchema;
8.zookeeper
mkdir /usr/soft/hbase
cp zoo_sample.cfg zoo.cfg

hadoop fs -mkdir -p /hbase
zkServer.sh start
zkCli.sh
create /hbase myhbase
9.hbase
vi hbase-env.sh

hbase-site.xml文件

<property>
	 <name>>hbase.rootdir</name>
	 <value>hdfs://love:9000/hbase</value>
</property>

hbase shell