关于hadoop:hadoop教程hdfs客户端开发

38次阅读

共计 5523 个字符，预计需要花费 14 分钟才能阅读完成。

hadoop 环境

hadoop 版本为

hadoop version
Hadoop 2.10.1
Subversion https://github.com/apache/hadoop -r 1827467c9a56f133025f28557bfc2c562d78e816
Compiled by centos on 2020-09-14T13:17Z
Compiled with protoc 2.5.0
From source with checksum 3114edef868f1f3824e7d0f68be03650

客户端开发

  • 引入依赖（应用 maven）
<dependency>
    <groupId>org.apache.hadoop</groupId>
    <artifactId>hadoop-client</artifactId>
    <version>2.10.1</version>
</dependency>

<dependency>
    <groupId>org.apache.hadoop</groupId>
    <artifactId>hadoop-hdfs</artifactId>
    <version>2.10.1</version>
</dependency>

<dependency>
    <groupId>org.apache.hadoop</groupId>
    <artifactId>hadoop-common</artifactId>
    <version>2.10.1</version>
</dependency>
  • 编写代码
package com.definesys.hadoop;

import org.apache.commons.io.IOUtils;
import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.hdfs.DistributedFileSystem;

import java.io.FileInputStream;
import java.io.IOException;


/**
 * @Description:
 * @author: jianfeng.zheng
 * @since: 2020/12/14 12:36 上午
 * @history: 1.2020/12/14 created by jianfeng.zheng
 */
public class HDFS {public static void main(String[] cmd) throws IOException {Configuration conf = new Configuration();
        conf.set("fs.defaultFS", "hdfs://master:9000/");
//        conf.set("fs.hdfs.impl", DistributedFileSystem.class.getName());
        System.setProperty("HADOOP_USER_NAME", "hadoop");
        FileSystem fs = FileSystem.get(conf);
        Path dst = new Path("hdfs://master:9000/demo/hello.txt");
        FSDataOutputStream os = fs.create(dst);
        FileInputStream is = new FileInputStream("/root/hello.txt");
        IOUtils.copy(is, os);
        is.close();
        os.close();
        fs.close();}
}
  • 打包

如果是 web 利用，个别会打包为 war 或者 ear，不论是哪种，这两种包格局都会把依赖包打进去，因而不必做非凡解决，如果须要本地运行，那么须要借助两个插件，把以下配置信息复制到 pom.xml 中

<build>
    <plugins>
        <plugin>
            <groupId>org.apache.maven.plugins</groupId>
            <artifactId>maven-jar-plugin</artifactId>
            <version>2.6</version>
            <configuration>
                <archive>
                    <manifest>
                        <addClasspath>true</addClasspath>
                        <classpathPrefix>lib/</classpathPrefix>
                        <mainClass>com.definesys.hadoop.HDFS</mainClass>
                    </manifest>
                </archive>
            </configuration>
        </plugin>

        <plugin>
            <groupId>org.apache.maven.plugins</groupId>
            <artifactId>maven-dependency-plugin</artifactId>
            <executions>
                <execution>
                    <id>copy-dependencies</id>
                    <phase>package</phase>
                    <goals>
                        <goal>copy-dependencies</goal>
                    </goals>
                    <configuration>
                        <outputDirectory>${project.build.directory}/lib</outputDirectory>
                        <overWriteReleases>false</overWriteReleases>
                        <overWriteSnapshots>false</overWriteSnapshots>
                        <overWriteIfNewer>true</overWriteIfNewer>
                    </configuration>
                </execution>
            </executions>
        </plugin>
    </plugins>
</build>

maven-jar-plugin 会依据配置生成 MANIFEST.MF 文件，MANIFEST.MF 文件记录运行类信息，依赖信息，相似以下这样

Manifest-Version: 1.0
Archiver-Version: Plexus Archiver
Built-By: asan
Class-Path: lib/hadoop-client-2.10.1.jar ....
Created-By: Apache Maven 3.6.3
Build-Jdk: 1.8.0_161
Main-Class: com.definesys.hadoop.HDFS

classpathPrefix 指定了依赖 jar 包所在的门路为 lib，maven-dependency-plugin 插件负责将依赖包全副 copy 到指定门路下，这里指定了 ${project.build.directory}/lib 目录，和 classpathPrefix 对应，打包实现后执行以下命令即可

java -jar hadoop-hdfs-1.0.jar

#或者手动指定运行类

java -cp hadoop-hdfs-1.0.jar com.definesys.hadoop.HDFS

打包还有一个插件 maven-assembly-plugin，不倡议应用这个插件进行打包，起因是这个插件会将所有依赖解压放到一个 jar 包里，hadoop 有些机制是通过 spi 实现，解压后会造成配置文件笼罩的状况

一个简略的 HDFS 操作类

package com.definesys.hadoop;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.Path;

import java.io.FileInputStream;
import java.io.IOException;


/**
 * @Description:
 * @author: jianfeng.zheng
 * @since: 2020/12/14 12:36 上午
 * @history: 1.2020/12/14 created by jianfeng.zheng
 */
public class HDFS {public static void main(String[] cmd) throws IOException {HDFS hdfs = new HDFS();
        hdfs.mkdir("/hdfsDemo");
        hdfs.putFile("/root/hello.txt", "/hdfsDemo");
        hdfs.dowloadFile("/hdfsDemo/hello.txt", "/root/hello-hdfs.txt");
        hdfs.deleteFile("/hdfsDemo");
    }

    public boolean mkdir(String path) throws IOException {FileSystem fs = this.getHDFSFileSystem();
        return fs.mkdirs(new Path(path));
    }

    public void putFile(String localPath, String hdfsPath) throws IOException {this.getHDFSFileSystem().copyFromLocalFile(new Path(localPath), new Path(hdfsPath));
    }

    public void deleteFile(String path) throws IOException {this.getHDFSFileSystem().delete(new Path(path), true);
    }

    public void dowloadFile(String hdfsPath, String localPath) throws IOException {this.getHDFSFileSystem().copyToLocalFile(new Path(hdfsPath), new Path(localPath));
    }

    private FileSystem getHDFSFileSystem() {Configuration conf = new Configuration();
        conf.set("fs.defaultFS", "hdfs://master:9000/");
        System.setProperty("HADOOP_USER_NAME", "hadoop");
        try {FileSystem fs = FileSystem.get(conf);
            return fs;
        } catch (IOException e) {throw new RuntimeException(e);
        }
    }
}

问题

权限问题

Exception in thread "main" org.apache.hadoop.security.AccessControlException: Permission denied: user=root, access=WRITE, inode="/":hadoop:supergroup:drwxr-xr-x
        at org.apache.hadoop.hdfs.server.namenode.FSPermissionChecker.check(FSPermissionChecker.java:350)
        at org.apache.hadoop.hdfs.server.namenode.FSPermissionChecker.checkPermission(FSPermissionChecker.java:251)
        at org.apache.hadoop.hdfs.server.namenode.FSPermissionChecker.checkPermission(FSPermissionChecker.java:189)

HDFS 文件系统权限和 Linux 相似，不同的用户对文件操作权限不一样，如果代码中没有指定用户名，那么就用执行程序的操作系统作为用户名，在这里是 root，咱们能够看下 hdfs 的文件权限

$ hadoop fs -ls /
Found 5 items
drwxr-xr-x   - asan   supergroup          0 2020-12-16 10:07 /001
drwx-w----   - hadoop supergroup          0 2020-12-07 10:54 /tmp
drwxr-xr-x   - hadoop supergroup          0 2020-12-07 11:05 /user

# 根门路权限

$ hadoop fs -ls -d /
drwxr-xr-x   - hadoop supergroup          0 2020-12-18 00:42 /

有几个解决方案

  • 批改根门路权限或者其余文件夹权限为 777
$ hadoop fs -chmod 777 /demo

$ hadoop fs -ls -d /demo
drwxrwxrwx   - hadoop supergroup          0 2020-12-18 00:46 /demo
  • 勾销权限验证

在 master 节点退出以下配置

<property>
    <name>dfs.permissions.enabled</name>
    <value>false</value>
</property>
  • 在代码中退出用户名配置（举荐）
System.setProperty("HADOOP_USER_NAME", "hadoop");

代码需在执行 hdfs 操作之前退出

正文完
hadoop
发表至： hadoop
2020-12-18
 0
Windows安装Hadoop
关于hadoop:深入浅出-Yarn-架构与实现41-ResourceManager-功能概述
hive小结
关于hadoop:初识Hadoop架构