Hadoop(一)MapReduce demo
Mapreduce基础编程模型:将一个大任务拆分成一个个小任务,再进行汇总。
MapReduce是分两个阶段:map阶段:拆;reduce阶段:聚合。
hadoop环境安装
安装:
1、解压 : tar -zxvf hadoop-2.4.1.tar.gz -C /root/training/
2、设置环境变量: vi ~/.bash_profile
HADOOP_HOME=/root/training/hadoop-2.7.3
export HADOOP_HOME
PATH=$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$PATH
export PATH
生效环境变量: source ~/.bash_profile
第一节:Hadoop的目录结构
第二节:Hadoop的本地模式
1、特点:不具备HDFS,只能测试MapReduce程序
2、修改hadoop-env.sh(echo $JAVA_HOME查出jdk安装路径:xx,将export JAVA_HOME=${JAVA_HOME}替换成export JAVA_HOME=xx)
修改第25行:export JAVA_HOME=/usr/java/jdk8u202-b08(行号可通过:esc后再set number来显示)
3、演示Demo: $HADOOP_HOME/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.7.3.jar
命令:hadoop jar hadoop-mapreduce-examples-2.7.3.jar wordcount ~/data/hadoop/input/test.txt ~/data/hadoop/output/wc
日志:19/09/16 10:45:00 INFO mapreduce.Job: map 100% reduce 100%
结果查看:
cd ~/data/hadoop/output/
ls
(前者是运行的结果集,后者是执行程序的状态)
more part-r-00000
注意:MR有一个默认的排序规则
第三节:Hadoop的伪分布模式
1、特点:具备Hadoop的所有功能,在单机上模拟一个分布式的环境
(1)HDFS:主:NameNode,数据节点:DataNode
(2)Yarn:容器,运行MapReduce程序
主节点:ResourceManager
从节点:NodeManager
2、步骤:
(1)hdfs-site.xml
<!--配置HDFS的冗余度-->
<property>
<name>dfs.replication</name>
<value>1</value>
</property>
<!--配置是否检查权限-->
<property>
<name>dfs.permissions</name>
<value>false</value>
</property>
(2)core-site.xml
<!--配置HDFS的NameNode-->
<property>
<name>fs.defaultFS</name>
<value>hdfs://192.168.88.11:9000</value>
</property>
<!--配置DataNode保存数据的位置-->
<property>
<name>hadoop.tmp.dir</name>
<value>/root/training/hadoop-2.7.3/tmp</value>
</property>
(3) mapred-site.xml
<!--配置MR运行的框架-->
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
(4) yarn-site.xml
<!--配置ResourceManager的地址-->
<property>
<name>yarn.resourcemanager.hostname</name>
<value>192.168.88.11</value>
</property>
<!--配置NodeManager执行任务的方式-->
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
(5) 格式化NameNode
hdfs namenode -format
日志:Storage directory /root/training/hadoop-2.7.3/tmp/dfs/name has been successfully formatted.
(6) 启动:start-all.sh
(*) HDFS: 存储数据
(*) Yarn:执行计算
(7) 访问:(*)命令行
(*)Java API
(*)Web Console:
HDFS:http://192.168.88.11:50070
Yarn:http://192.168.88.11:8088
到这里已经能够通过外部访问了
web console无法通过http://ip:port访问服务页面问题排查
原文出自(https://blog.csdn.net/hanwenshan123/article/details/78717782)
问题1:hdfs-site.xml配置项
通过jps命令查看java进程的状态,HADOOP相关的进程运行正常。(jps是jdk提供的一个查看当前java进程的小工具, 可以看做是JavaVirtual Machine Process Status Tool的缩写)
[root@node4 ~]# jps
25059 SecondaryNameNode
25347 ResourceManager
25556 NodeManager
24805 DataNode
29269 Jps
24633 NameNode
通过netstat命令查看网络端口服务情况,发现local address列给出的ip地址除了127.0.0.1就是0.0.0.0,这些本地有效的地址,是无法对外提供服务的,这才是问题的关键。
[root@node4 ~]# netstat -ntlp
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp 0 0 127.0.0.1:43759 0.0.0.0:* LISTEN 24805/java
tcp 0 0 0.0.0.0:50070 0.0.0.0:* LISTEN 24633/java
tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN 12782/sshd
tcp 0 0 127.0.0.1:25 0.0.0.0:* LISTEN 2325/master
tcp 0 0 0.0.0.0:50010 0.0.0.0:* LISTEN 24805/java
tcp 0 0 0.0.0.0:50075 0.0.0.0:* LISTEN 24805/java
tcp 0 0 0.0.0.0:50020 0.0.0.0:* LISTEN 24805/java
tcp 0 0 127.0.0.1:9000 0.0.0.0:* LISTEN 24633/java
tcp 0 0 0.0.0.0:50090 0.0.0.0:* LISTEN 25059/java
tcp6 0 0 :::22 :::* LISTEN 12782/sshd
tcp6 0 0 127.0.0.1:8088 :::* LISTEN 25347/java
tcp6 0 0 ::1:25 :::* LISTEN 2325/master
tcp6 0 0 :::13562 :::* LISTEN 25556/java
tcp6 0 0 :::43451 :::* LISTEN 25556/java
tcp6 0 0 127.0.0.1:8030 :::* LISTEN 25347/java
tcp6 0 0 127.0.0.1:8031 :::* LISTEN 25347/java
tcp6 0 0 127.0.0.1:8032 :::* LISTEN 25347/java
tcp6 0 0 127.0.0.1:8033 :::* LISTEN 25347/java
tcp6 0 0 :::8040 :::* LISTEN 25556/java
tcp6 0 0 :::8042 :::* LISTEN 25556/java
修改HADOOP_HOME/etc/hadoop/hdfs-site.xml文件,加入
<property>
<name>dfs.namenode.http-address</name>
<value>node4:50070</value>
</property>
或者加入
<property>
<name>dfs.namenode.http-address</name>
<value>hdfs://192.168.88.11:50070</value>
</property>
再次用netstat -ntlp查看
[root@node4 ~]# netstat -ntlp
Active Internet connections (only servers)
Proto Recv-Q Send-Q Local Address Foreign Address State PID/Program name
tcp 0 0 127.0.0.1:43759 0.0.0.0:* LISTEN 24805/java
tcp 0 0 10.60.8.28.50070 0.0.0.0:* LISTEN 24633/java
tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN 12782/sshd
tcp 0 0 127.0.0.1:25 0.0.0.0:* LISTEN 2325/master
tcp 0 0 0.0.0.0:50010 0.0.0.0:* LISTEN 24805/java
tcp 0 0 0.0.0.0:50075 0.0.0.0:* LISTEN 24805/java
tcp 0 0 0.0.0.0:50020 0.0.0.0:* LISTEN 24805/java
tcp 0 0 127.0.0.1:9000 0.0.0.0:* LISTEN 24633/java
tcp 0 0 0.0.0.0:50090 0.0.0.0:* LISTEN 25059/java
tcp6 0 0 :::22 :::* LISTEN 12782/sshd
tcp6 0 0 127.0.0.1:8088 :::* LISTEN 25347/java
tcp6 0 0 ::1:25 :::* LISTEN 2325/master
tcp6 0 0 :::13562 :::* LISTEN 25556/java
tcp6 0 0 :::43451 :::* LISTEN 25556/java
tcp6 0 0 127.0.0.1:8030 :::* LISTEN 25347/java
tcp6 0 0 127.0.0.1:8031 :::* LISTEN 25347/java
tcp6 0 0 127.0.0.1:8032 :::* LISTEN 25347/java
tcp6 0 0 127.0.0.1:8033 :::* LISTEN 25347/java
tcp6 0 0 :::8040 :::* LISTEN 25556/java
tcp6 0 0 :::8042 :::* LISTEN 25556/java
问题2:selinux
按照道理应该可以访问50070端口了,但是仍然不行。再检查selinux,发现状态是enabled。
- 查看SELINUX的状态
[root@node4 ~]# /usr/sbin/sestatus -v
SELinux status: enabled
SELinuxfs mount: /sys/fs/selinux
SELinux root directory: /etc/selinux
Loaded policy name: targeted
Current mode: enforcing
Mode from config file: enforcing
Policy MLS status: enabled
Policy deny_unknown status: allowed
Max kernel policy version: 28
Process contexts:
Current context: unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023
Init context: system_u:system_r:init_t:s0
/usr/sbin/sshd system_u:system_r:sshd_t:s0-s0:c0.c1023
File contexts:
Controlling terminal: unconfined_u:object_r:user_devpts_t:s0
/etc/passwd system_u:object_r:passwd_file_t:s0
/etc/shadow system_u:object_r:shadow_t:s0
/bin/bash system_u:object_r:shell_exec_t:s0
/bin/login system_u:object_r:login_exec_t:s0
/bin/sh system_u:object_r:bin_t:s0 -> system_u:object_r:shell_exec_t:s0
/sbin/agetty system_u:object_r:getty_exec_t:s0
/sbin/init system_u:object_r:bin_t:s0 -> system_u:object_r:init_exec_t:s0
/usr/sbin/sshd system_u:object_r:sshd_exec_t:s0
编辑/etc/selinux/config文件SELINUX=enforcing修改成SELINUX=disable,重启服务器。再试。修改后的selinux
[root@node4 ~]# /usr/sbin/sestatus -v
SELinux status: disabled
问题3:firewall(iptables端口开放)
关闭selinux之后,仍然无法访问页面,再查看iptables防火墙的设置
[root@node4 sbin]# firewall-cmd --state
running
[root@node4 sbin]# firewall-cmd --get-service
RH-Satellite-6 amanda-client amanda-k5-client bacula bacula-client bitcoin bitcoin-rpc bitcoin-testnet bitcoin-testnet-rpc ceph ceph-mon cfengine condor-collector ctdb dhcp dhcpv6 dhcpv6-client dns docker-registry dropbox-lansync elasticsearch freeipa-ldap freeipa-ldaps freeipa-replication freeipa-trust ftp ganglia- client ganglia-master high-availability http https imap imaps ipp ipp-client ipsec iscsi-target kadmin kerberos kibana klogin kpasswd kshell ldap ldaps libvirt libvirt-tls managesieve mdns mosh mountd ms-wbt mssql mysql nfs nrpe ntp openvpn ovirt-imageio ovirt-storageconsole ovirt-vmconsole pmcd pmproxy pmwebapi pmwebapis pop3 pop3s postgresql privoxy proxy-dhcp ptp pulseaudio puppetmaster quassel radius rpc-bind rsh rsyncd samba samba-client sane sip sips smtp smtp-submission smtps snmp snmptrap spideroak-lansync squid ssh synergy syslog syslog-tls telnet tftp tftp-client tinc tor-socks transmission-client vdsm vnc-server wbem-https xmpp-bosh xmpp-client xmpp-local xmpp-server
增加50070端口到允许,重启防火墙服务
[root@node4 sbin]# firewall-cmd --zone=public --add-port=50070/tcp --permanent
success
[root@node4 sbin]# firewall-cmd --reload
success
处理结果
问题4.8088端口无法访问yarn
修改yarn-site.xml文件,在<configuration></configuration>添加:
<property>
<name>yarn.resourcemanager.webapp.address</name>
<value>192.168.88.11:8088</value>
</property>
集群分布模式
1.将hadoop整个安装目录拷贝到其他两台机器
scp -r /home/xxxx/hadoop XXX@hadoop02:/home/xxxx/
scp -r /home/xxxx/hadoop XXX@hadoop03:/home/xxxx/
2.修改主机上的slaves文件内容为从节点的主机名称:
hadoop1
hadoop2
hadoop3