Docker部署Hadoop-03-Docker部署Hadoop

参考

Docker部署Hadoop-01-Docker安装

Docker部署Hadoop-02-Docker常见操作

Ubuntu使用Docker部署搭建Hadoop集群

一、环境准备

# 更新系统
sudo apt update && sudo apt upgrade -y

# 安装Docker
sudo apt install docker.io -y #安装不了要换（清华、阿里等）源
sudo systemctl start docker #现在启动
sudo systemctl enable docker #开机自启动

# 验证Docker安装
docker –version

# 创建专用目录
mkdir ~/hadoop-docker && cd ~/hadoop-docker

准备一个 hadoop-3.3.0.tar.gz

二、创建Docker网络

Hadoop集群需要容器间通信，我们先创建专用网络：

docker network create \\
–driver bridge \\
–subnet=172.19.0.0/16 \\
hadoop-net

验证网络情况

docker network inspect hadoop-net

三、构建基础 镜像

创建启动脚本：

sudo nano entrypoint.sh

内容：

#!/bin/bash

# 启动SSH服务
service ssh start

# 生成SSH密钥（如果不存在）
if [ ! -f ~/.ssh/id_rsa ]; then
ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa
cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys
chmod 600 ~/.ssh/authorized_keys
fi

# 保持容器运行
tail -f /dev/null

创建Dockerfile：

sudo nano Dockerfile

内容如下：

FROM ubuntu:22.04

# 设置时区避免交互提示
ENV DEBIAN_FRONTEND=noninteractive

# 安装基础工具
RUN apt update && apt install -y \\
openssh-server \\
openjdk-11-jdk \\
wget \\
vim \\
net-tools \\
iputils-ping \\
dnsutils

# 配置SSH
RUN mkdir /var/run/sshd
RUN echo 'root:root' | chpasswd
RUN sed -i 's/#PermitRootLogin prohibit-password/PermitRootLogin yes/' /etc/ssh/sshd_config
RUN sed -i 's/#PasswordAuthentication yes/PasswordAuthentication yes/' /etc/ssh/sshd_config

# 安装Hadoop
#ENV HADOOP_VERSION=3.4.1
#ENV HADOOP_URL=https://downloads.apache.org/hadoop/common/hadoop-$HADOOP_VERSION/hadoop-$HADOOP_VERSION.tar.gz

#RUN wget $HADOOP_URL -O hadoop.tar.gz && \\
# tar -xzvf hadoop.tar.gz -C /usr/local/ && \\
# rm hadoop.tar.gz && \\
# mv /usr/local/hadoop-$HADOOP_VERSION /usr/local/hadoop
# 复制本地Hadoop安装包（核心修改：不再从网上下载）
ENV HADOOP_VERSION=3.3.0
COPY ./hadoop-$HADOOP_VERSION.tar.gz /tmp/

# 安装Hadoop（使用本地包）
RUN tar -xzvf /tmp/hadoop-$HADOOP_VERSION.tar.gz -C /usr/local/ \\
&& rm /tmp/hadoop-$HADOOP_VERSION.tar.gz \\
&& mv /usr/local/hadoop-$HADOOP_VERSION /usr/local/hadoop \\
&& chown -R root:root /usr/local/hadoop

# 设置环境变量
ENV JAVA_HOME=/usr/lib/jvm/java-11-openjdk-amd64
ENV HADOOP_HOME=/usr/local/hadoop
ENV HADOOP_MAPRED_HOME=/usr/local/hadoop # 新增这一行
ENV PATH=$PATH:$HADOOP_HOME/bin:$HADOOP_HOME/sbin

# 创建数据目录
RUN mkdir -p /usr/local/hadoop/namenode_dir
RUN mkdir -p /usr/local/hadoop/datanode_dir
RUN mkdir -p /usr/local/hadoop/tmp

# 暴露端口
EXPOSE 22 9870 8088 9000 50070 50010

# 启动脚本
COPY entrypoint.sh /entrypoint.sh
RUN chmod +x /entrypoint.sh
ENTRYPOINT ["/entrypoint.sh"]

构建镜像：

docker build -t hadoop-base .

如果需要删除镜像

构建镜像：

```bash
docker rmi hadoop-base:latest

## 四、启动Hadoop集群

1. 启动Master节点

```bash
docker run -itd \\
–name master \\
–hostname master \\
–net hadoop-net \\
–ip 172.19.0.2 \\
-p 9870:9870 \\
-p 8088:8088 \\
-p 9000:9000 \\
hadoop-base

docker run -itd \\
–name worker01 \\
–hostname worker01 \\
–net hadoop-net \\
–ip 172.19.0.3 \\
hadoop-base

docker run -itd \\
–name worker02 \\
–hostname worker02 \\
–net hadoop-net \\
–ip 172.19.0.4 \\
hadoop-base

五、配置主机解析

在Master节点中：

docker exec master bash -c "echo '172.19.0.2 master' >> /etc/hosts"
docker exec master bash -c "echo '172.19.0.3 worker01' >> /etc/hosts"
docker exec master bash -c "echo '172.19.0.4 worker02' >> /etc/hosts"

在Worker01节点中：

docker exec worker01 bash -c "echo '172.19.0.2 master' >> /etc/hosts"
docker exec worker01 bash -c "echo '172.19.0.3 worker01' >> /etc/hosts"
docker exec worker01 bash -c "echo '172.19.0.4 worker02' >> /etc/hosts"

在Worker02节点中：

docker exec worker02 bash -c "echo '172.19.0.2 master' >> /etc/hosts"
docker exec worker02 bash -c "echo '172.19.0.3 worker01' >> /etc/hosts"
docker exec worker02 bash -c "echo '172.19.0.4 worker02' >> /etc/hosts"

六、配置SSH免密登录

进入Master容器：

docker exec -it master bash

生成SSH密钥：这一个步骤不用执行，已经执行过了

ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa

配置免密登录：

# 将公钥复制到所有节点
ssh-copy-id master
ssh-copy-id worker01
ssh-copy-id worker02

# 测试免密登录
ssh worker01 hostname # 应返回 worker01
ssh worker02 hostname # 应返回 worker02

如果 ssh-copy-id worker01 出现错误，可以按照下面的解决方案

# 这个是命令说明不用执行
ssh-copy-id master
# 等价
# 手动复制公钥（替代 ssh-copy-id master）
cat ~/.ssh/id_rsa.pub | ssh root@master "mkdir -p ~/.ssh && cat >> ~/.ssh/authorized_keys && chmod 600 ~/.ssh/authorized_keys && chmod 700 ~/.ssh"

步骤 1：先登录 worker01 容器，检查并修复 SSH 配置

# 从宿主机直接进入 worker01 容器（绕开 SSH，直接操作）
docker exec -it worker01 bash

# 1. 重置 root 密码为 root（确保密码正确）
echo 'root:root' | chpasswd

# 2. 确认 SSH 配置中开启密码登录（关键）
sed -i 's/#PermitRootLogin prohibit-password/PermitRootLogin yes/' /etc/ssh/sshd_config
sed -i 's/#PasswordAuthentication yes/PasswordAuthentication yes/' /etc/ssh/sshd_config

# 3. 重启 SSH 服务，让配置生效
service ssh restart

# 4. 退出 worker01 容器
exit

步骤 2：重新执行 ssh-copy-id worker01（验证）

回到 master 容器的终端，重新执行命令：

# 在 master 容器内执行
ssh-copy-id worker01

• 此时会提示 Are you sure you want to continue connecting (yes/no/[fingerprint])?，输入 yes；
• 接着提示 root@worker01's password:，输入 root（你重置后的密码），回车即可成功。

步骤 3：同理修复 worker02（如果有）

# 宿主机执行：进入 worker02 容器
docker exec -it worker02 bash

# 重置密码+开启SSH密码登录
echo 'root:root' | chpasswd
sed -i 's/#PermitRootLogin prohibit-password/PermitRootLogin yes/' /etc/ssh/sshd_config
sed -i 's/#PasswordAuthentication yes/PasswordAuthentication yes/' /etc/ssh/sshd_config
service ssh restart
exit

# master 容器内执行：复制公钥到 worker02
ssh-copy-id worker02

步骤 4：验证免密登录是否生效

# 在 master 容器内测试
ssh worker01 # 应直接登录，无密码提示
ssh worker02 # 同理

七、配置Hadoop集群

mkdir /usr/local/hadoop/etc/hadoop/config
cd /usr/local/hadoop/etc/hadoop/config

vi core-site.xml

内容：

<configuration>
<property>
<name>fs.defaultFS</name>
<value>hdfs://master:9000</value>
</property>
<property>
<name>hadoop.tmp.dir</name>
<value>/usr/local/hadoop/tmp</value>
</property>
</configuration>

vi hdfs-site.xml

内容：

<configuration>
<property>
<name>dfs.replication</name>
<value>2</value>
</property>
<property>
<name>dfs.namenode.name.dir</name>
<value>/usr/local/hadoop/namenode_dir</value>
</property>
<property>
<name>dfs.datanode.data.dir</name>
<value>/usr/local/hadoop/datanode_dir</value>
</property>
<property>
<name>dfs.permissions</name>
<value>false</value>
</property>
</configuration>

vi mapred-site.xml

内容：

<configuration>
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
<property>
<name>yarn.app.mapreduce.am.env</name>
<value>HADOOP_MAPRED_HOME=$HADOOP_HOME</value>
</property>
<property>
<name>mapreduce.map.env</name>
<value>HADOOP_MAPRED_HOME=$HADOOP_HOME</value>
</property>
<property>
<name>mapreduce.reduce.env</name>
<value>HADOOP_MAPRED_HOME=$HADOOP_HOME</value>
</property>
</configuration>

vi yarn-site.xml

内容：

<configuration>
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
<property>
<name>yarn.resourcemanager.hostname</name>
<value>master</value>
</property>
<property>
<name>yarn.nodemanager.env-whitelist</name>
<value>JAVA_HOME,HADOOP_COMMON_HOME,HADOOP_HDFS_HOME,HADOOP_CONF_DIR,CLASSPATH_PREPEND_DISTCACHE,HADOOP_YARN_HOME,HADOOP_MAPRED_HOME</value>
</property>
<property>
<name>yarn.application.classpath</name>
<value>
$HADOOP_HOME/etc/hadoop,
$HADOOP_HOME/share/hadoop/common/*,
$HADOOP_HOME/share/hadoop/common/lib/*,
$HADOOP_HOME/share/hadoop/hdfs/*,
$HADOOP_HOME/share/hadoop/hdfs/lib/*,
$HADOOP_HOME/share/hadoop/mapreduce/*,
$HADOOP_HOME/share/hadoop/mapreduce/lib/*,
$HADOOP_HOME/share/hadoop/yarn/*,
$HADOOP_HOME/share/hadoop/yarn/lib/*
</value>
</property>
</configuration>

vi workers

内容：

worker01
worker02

vi /usr/local/hadoop/etc/hadoop/hadoop-env.sh

添加

export JAVA_HOME=/usr/lib/jvm/java-11-openjdk-amd64
export HADOOP_HOME=/usr/local/hadoop
export HADOOP_CONF_DIR=$HADOOP_HOME/etc/hadoop
export HDFS_NAMENODE_USER=root
export HDFS_DATANODE_USER=root
export HDFS_SECONDARYNAMENODE_USER=root
export YARN_RESOURCEMANAGER_USER=root
export YARN_NODEMANAGER_USER=root

# 在Master容器中执行
for worker in worker01 worker02; do
scp -r $HADOOP_HOME/etc/hadoop/config/* $worker:$HADOOP_HOME/etc/hadoop/
scp $HADOOP_HOME/etc/hadoop/hadoop-env.sh $worker:$HADOOP_HOME/etc/hadoop/
done

在把master容器中的文件覆盖到/etc/hadoop中

# 进入 Hadoop 配置根目录
cd /usr/local/hadoop/etc/hadoop/

# 强制覆盖：将 config 目录下的所有配置文件复制到当前目录（覆盖原有文件）
cp -f config/* ./

# 验证覆盖结果（查看核心文件是否已更新）
ls -l core-site.xml hdfs-site.xml yarn-site.xml mapred-site.xml workers

八、启动Hadoop集群

hdfs namenode -format

成功标志：看到Storage directory has been successfully formatted信息

如果失败，需要重新格式化，格式化前，删除原有的数据

# 清空 NameNode 元数据目录（namenode_dir）
rm -rf /usr/local/hadoop/namenode_dir/*

# 清空 DataNode 数据目录（datanode_dir）
rm -rf /usr/local/hadoop/datanode_dir/*

# 清空 Hadoop 临时目录（tmp）
rm -rf /usr/local/hadoop/tmp/*

# 验证清空结果（目录为空，输出 total 0）
echo "=== 验证目录清空结果 ==="
ls -l /usr/local/hadoop/namenode_dir/
ls -l /usr/local/hadoop/datanode_dir/
ls -l /usr/local/hadoop/tmp/

start-dfs.sh

验证：

jps

应看到： xxxxx NameNode xxxxx DataNode xxxxx SecondaryNameNode

start-yarn.sh

验证

jps

九、验证集群功能

hdfs dfs -mkdir -p /user/root/input

echo "Hello Hadoop World" > test.txt
hdfs dfs -put test.txt /user/root/input

hdfs dfs -ls /user/root/input

hadoop jar $HADOOP_HOME/share/hadoop/mapreduce/hadoop-mapreduce-examples-3.3.0.jar wordcount /user/root/input /user/root/output

如果出现问题，可以强制执行

# 终极兜底命令（复制粘贴直接运行）
hadoop jar \\
$HADOOP_HOME/share/hadoop/mapreduce/hadoop-mapreduce-examples-3.3.0.jar \\
wordcount \\
-Dmapreduce.application.classpath=$(hadoop classpath) \\
-Dyarn.app.mapreduce.am.env="HADOOP_MAPRED_HOME=/usr/local/hadoop" \\
-Dmapreduce.map.env="HADOOP_MAPRED_HOME=/usr/local/hadoop" \\
-Dmapreduce.reduce.env="HADOOP_MAPRED_HOME=/usr/local/hadoop" \\
/user/root/input \\
/user/root/output

hdfs dfs -cat /user/root/output/part-r-00000

应看到：

重新设置全局的环境变量，在master中，这个方法也没有生效

echo "export HADOOP_MAPRED_HOME=/usr/local/hadoop" >> /etc/profile
source /etc/profile
# 同步到 worker 节点
for node in worker01 worker02; do
ssh $node "echo 'export HADOOP_MAPRED_HOME=/usr/local/hadoop' >> /etc/profile && source /etc/profile"
done