java写hdfs程序

1、hadoop默认临时数据文件是存储于Unix的tmp目录下（cd /tmp 包含hadoop-root等文件），如果不进行修改，linux系统重启后hadoop有可能出现不正常现象；故需要修改hadoop的临时文件存放目录
2、vim core-site.xml 配置如下，然后重启hadoop集群，不要对namenode重新进行格式化操作
修改datanode /var/hadoop/dfs/data/current 目录下VERSION文件的clusterid与namenode一致；然后启动集群正常

hadoop.tmp.dir
/var/hadoop

在namenode执行格式化操作后，会导致namenode重新生成clusterid，而datanode的clusterID值没变，
namenode与datanode clusterid不一致导致datanode启动异常；需要手动改成与namenode一致
3、测试时，可以关闭权限检查(否则没有权限访问)，在namenode节点添加如下配置
vim hdfs-site.xml

dfs.permissions.enabled
false

xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
4.0.0
com.skcc
wordcount
0.0.1-SNAPSHOT
wordcount
count the word

    UTF-8    2.7.3            junit        junit        4.12                org.apache.hadoop        hadoop-client        ${hadoop.version}                org.apache.hadoop        hadoop-common        ${hadoop.version}                org.apache.hadoop        hadoop-hdfs        ${hadoop.version}    

package com.skcc.hadoop;

import java.io.FileInputStream;
import java.io.IOException;
import java.io.InputStream;
import java.net.URL;
import java.text.NumberFormat;

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.FSDataOutputStream;
import org.apache.hadoop.fs.FileSystem;
import org.apache.hadoop.fs.FsUrlStreamHandlerFactory;
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.io.IOUtils;

public class HelloHDFS {

public HelloHDFS() {    // TODO Auto-generated constructor stub}public static FileSystem getFileSystemInstance() {    Configuration conf = new Configuration();    conf.set("fs.defaultFS", "hdfs://172.26.19.40:9000");    FileSystem fileSystem = null;    try {        fileSystem = FileSystem.get(conf);    } catch (IOException e) {        // TODO Auto-generated catch block        e.printStackTrace();    }    return fileSystem;}public static void getFileFromHDFS() throws Exception {    //URL 默认处理http协议， FsUrlStreamHandlerFactory 处理hdfs协议    URL.setURLStreamHandlerFactory(new FsUrlStreamHandlerFactory());    URL url=new URL("hdfs://172.26.19.40:9000/10803060234.txt");    InputStream inputStream= url.openStream();    IOUtils.copyBytes(inputStream, System.out, 4096,true);}public static void getFileFromBaiDu() throws IOException {    URL url=new URL("http://skynet.skhynix-cq.com.cn/plusWare/Main.aspx");    InputStream inputStream= url.openStream();    IOUtils.copyBytes(inputStream, System.out, 4096,true);}public static void testHadoop() throws Exception {    FileSystem fileSystem = getFileSystemInstance();    Boolean success = fileSystem.mkdirs(new Path("/skcc"));    System.out.println("mkdirs is " + success);    success = fileSystem.exists(new Path("/10803060234.txt"));    System.out.println("file exists is " + success);    success = fileSystem.delete(new Path("/test2.data"),true);    System.out.println("delete dirs is " + success);    success = fileSystem.exists(new Path("/skcc"));    System.out.println("dirs exists is "+ success);}public static void uploadFileToHDFS() throws Exception {    FileSystem fileSystem = getFileSystemInstance();    String filename = "/test2.data";    // overwrite ==true    FSDataOutputStream outputStream = fileSystem.create(new Path(filename), true);    FileInputStream fis = new FileInputStream("D:\\2018\\u001.zip");

// IOUtils.copyBytes(fis, outputStream, 4096, true);

    long totalLen = fis.getChannel().size();    long tmpSize = 0;    double readPercent = 0;    NumberFormat numberFormat = NumberFormat.getInstance();    numberFormat.setMaximumFractionDigits(0);    System.out.println("totalLen : " + totalLen + " available : " + fis.available());    byte[] buf = new byte[4096];    int len = fis.read(buf);    while (len != -1) {        tmpSize = tmpSize + len;        String result = numberFormat.format((float)tmpSize / (float)totalLen * 100 );        outputStream.write(buf,0,len);        System.out.println("Upload Percent : " + result + "%");        len = fis.read(buf);    }}

}