• 伪分布模式下使用java接口,访问hdfs


     1 package com.bq.pro;
     2 
     3 import java.io.IOException;
     4 
     5 import org.apache.hadoop.conf.Configuration;
     6 import org.apache.hadoop.fs.FSDataInputStream;
     7 import org.apache.hadoop.fs.FSDataOutputStream;
     8 import org.apache.hadoop.fs.FileSystem;
     9 import org.apache.hadoop.fs.FileStatus;
    10 import org.apache.hadoop.fs.Path;
    11 import org.apache.hadoop.util.GenericOptionsParser;
    12 import org.apache.hadoop.util.Progressable;
    13 
    14 public class MergeFiles {
    15     @SuppressWarnings("deprecation")
    16     public static void main(String[] args) throws IOException {
    17         Configuration conf=new Configuration();
    18         FileSystem localInputFs=FileSystem.getLocal(conf);
    19         //FileSystem hdfsOutputFs=FileSystem.get(conf);
    20         String[] otherargs=new GenericOptionsParser(conf,args).getRemainingArgs();
    21         Path inputLocalDir=new Path(otherargs[0]);
    22         Path hdfsPathOutput=new Path(otherargs[1]);
    23         FileSystem hdfsOutputFs=hdfsPathOutput.getFileSystem(conf);
    24         try {
    25             FileStatus[] inputFiles=localInputFs.listStatus(inputLocalDir);
    26             FSDataOutputStream out=hdfsOutputFs.create(hdfsPathOutput,new Progressable() {
    27                 
    28                 @Override
    29                 public void progress() {
    30                     System.out.print(".");
    31                     
    32                 }
    33             });
    34             for(int i=0;i<inputFiles.length;i++)
    35             {
    36                 FSDataInputStream in=localInputFs.open(inputFiles[i].getPath());
    37                 byte[] buffer=new byte[100];
    38                 int byteRead=0;
    39                 while((byteRead=in.read(buffer))>0)
    40                 {
    41                     out.write(buffer, 0, byteRead);
    42                     
    43                 }
    44                 in.close();
    45             }
    46             out.close();
    47             //localInputFs.delete(inputLocalDir);
    48             
    49             
    50         } catch (IOException e) {
    51             // TODO: handle exception
    52             e.printStackTrace();
    53         }
    54     }
    55 
    56 }

    在参数中输入:/home/haduser/workspace/MergeCDFiles/resources/    hdfs://localhost:9000/opt/hadoop/tmp/test/a.txt

    伪分布模式下用FileSystem.get(conf) 获取hdfs会出现:

    Exception in thread "main" java.lang.IllegalArgumentException: Wrong FS: hdfs://localhost:9000/opt/hadoop/tmp/test, expected: file:///

        at org.apache.hadoop.fs.FileSystem.checkPath(FileSystem.java:390)
        at org.apache.hadoop.fs.RawLocalFileSystem.pathToFile(RawLocalFileSystem.java:55)
        at org.apache.hadoop.fs.RawLocalFileSystem.mkdirs(RawLocalFileSystem.java:340)
        at org.apache.hadoop.fs.ChecksumFileSystem.mkdirs(ChecksumFileSystem.java:492)
        at org.apache.hadoop.fs.ChecksumFileSystem.create(ChecksumFileSystem.java:377)
        at org.apache.hadoop.fs.ChecksumFileSystem.create(ChecksumFileSystem.java:364)
        at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:564)
        at org.apache.hadoop.fs.FileSystem.create(FileSystem.java:464)
        at com.bq.pro.MergeFiles.main(MergeFiles.java:26)

    使用FileSystem hdfsOutputFs=hdfsPathOutput.getFileSystem(conf);来代替的时候就不会用问题,具体原因不详,应该是伪分布下的问题,在集群里应该不会有问题。




  • 相关阅读:
    ThreadPoolExecutor
    HashMap
    ReentrantReadWriteLock
    AbstractQueuedSynchronizer
    KAFKA:如何做到1秒发布百万级条消息
    Kafka生产者性能优化之吞吐量VS延迟
    Hadoop YARN:调度性能优化实践(转)
    Windows下安装MySQL压缩zip包
    Android 8 蓝牙打开过程
    msm audio platform 驱动代码跟踪
  • 原文地址:https://www.cnblogs.com/wzyj/p/3437059.html
Copyright © 2020-2023  润新知