hadoop - HDFS has file but java.io.FileNotFoundException happens -

i running mapreduce program on hadoop.

the inputformat passes each file path mapper.

i can check file through cmd this,

$ hadoop fs -ls hdfs://slave1.kdars.com:8020/user/hadoop/num_5/13.pdf

found 1 items -rwxrwxrwx 3 hdfs hdfs 184269 2015-03-31 22:50 hdfs://slave1.kdars.com:8020/user/hadoop/num_5/13.pdf

however when try open file mapper side, not working.

15/04/01 06:13:04 info mapreduce.job: task id : attempt_1427882384950_0025_m_000002_2, status : failed error: java.io.filenotfoundexception: hdfs:/slave1.kdars.com:8020/user/hadoop/num_5/13.pdf (no such file or directory)

at java.io.fileinputstream.open(native method) @ java.io.fileinputstream.<init>(fileinputstream.java:146) @ java.io.fileinputstream.<init>(fileinputstream.java:101) @ org.apache.pdfbox.pdmodel.pddocument.load(pddocument.java:1111) 

i checked inputformat work fine , mapper have got right file path. mapper code this,

@override public void map(text title, text file, context context) throws ioexception, interruptedexception {      long time = system.currenttimemillis();      simpledateformat daytime = new simpledateformat("yyyy-mm-dd hh:mm:ss");     string str = daytime.format(new date(time));      file temp = new file(file.tostring());     if(temp.exists()){         dbmanager.getinstance().insertsql("insert `plagiarismdb`.`workflow` (`type`) value ('"+temp+" exists')");     }else{         dbmanager.getinstance().insertsql("insert `plagiarismdb`.`workflow` (`type`) value ('"+temp+" not exists')");     } } 

help me please.

try in mapper method. first import these. import org.apache.hadoop.fs.filesystem; import org.apache.hadoop.fs.path;

use in mapper method.

     filesystem fs = filesystem.get(new configuration());        path path=  new path(value.tostring());       system.out.println(path);        if(fs.exists(path))          context.write(value, one);          else              context.write(value, zero);