java - unexpected multiple execution of mapper intended to run once -

i tried write simple job 1 mapper , no reducer write data hbase. in mapper tried open connection hbase, write few rows of data table , close connection. in job driver using jobconf.setnummaptasks(1); , jobconf.setnumreducetasks(0); specify 1 mapper , no reducer executed. setting reducer class identityreducer in jobconf. strange behavior observing job writes data hbase table after see in logs continuously tried open connection hbase , closes connection goes on 20-30 minutes , after job declared have completed 100% success. @ end when check _success file created dummy data put in outputcollector.collect(...) see hundred of rows of dummy data when there should 1. following code job driver

    public int run(string[] arg0) throws exception {         configuration config = hbaseconfiguration.create(getconf());         ensurerequiredparametersexist(config);         ensureoptionalparametersexist(config);          jobconf jobconf = new jobconf(config, getclass());         jobconf.setjobname(config.get(etljobconstants.etl_job_name));         //set map specific configuration         jobconf.setnummaptasks(1);         jobconf.setmaxmapattempts(1);         jobconf.setinputformat(textinputformat.class);         jobconf.setmapperclass(singletonmapper.class);         jobconf.setmapoutputkeyclass(longwritable.class);         jobconf.setmapoutputvalueclass(text.class);          //set reducer specific configuration         jobconf.setreducerclass(identityreducer.class);         jobconf.setoutputkeyclass(longwritable.class);         jobconf.setoutputvalueclass(text.class);         jobconf.setoutputformat(textoutputformat.class);         jobconf.setnumreducetasks(0);          //set job specific configuration details input file name etc         fileinputformat.setinputpaths(jobconf, jobconf.get(etljobconstants.etl_job_file_input_path));         system.out.println("setting output path : " + jobconf.get(etljobconstants.etl_job_file_output_path));         fileoutputformat.setoutputpath(jobconf,                 new path(jobconf.get(etljobconstants.etl_job_file_output_path)));         jobclient.runjob(jobconf);         return 0;     } 

driver class extends configured , implements tool (i used sample definitive guide)following code in mapper class.

following code in mapper's map method open connection hbase, preliminary check make sure table exists , write rows , close table.

    public void map(longwritable arg0, text arg1,         outputcollector<longwritable, text> arg2, reporter arg3)         throws ioexception {       htable atable = null;     hbaseadmin admin = null;       try {          arg3.setstatus("started");          /*          * set-up hbase config          */         admin = new hbaseadmin(conf);          /*          * open connection table          */         string tablename = conf.get(etljobconstants.etl_job_table_name);          htabledescriptor htd = new htabledescriptor(tobytes(tablename));         string colfamilyname = conf.get(etljobconstants.etl_job_table_column_family_name);          byte[] tablename = htd.getname();         /* call function ensure table 'tablename' exists */          /*          * loop , put file data table          */         atable = new htable(conf, tablename);          datarow row = /* logic generate data */         while (row != null) {             byte[] rowkey = tobytes(row.getrowkey());             put put = new put(rowkey);             (datanode node : row.getrowdata()) {                 put.add(tobytes(colfamilyname), tobytes(node.getnodename()),                         tobytes(node.getnodevalue()));             }             atable.put(put);             arg3.setstatus("xoxoxoxoxoxoxoxoxoxoxoxo added data row hbase");             row = fileparser.getnextrow();         }         atable.flushcommits();         arg3.setstatus("xoxoxoxoxoxoxoxoxoxoxoxo finished adding data hbase");      } {         if (atable != null) {             atable.close();         }          if (admin != null) {             admin.close();         }     }      arg2.collect(new longwritable(10), new text("something"));     arg3.setstatus("xoxoxoxoxoxoxoxoxoxoxoxoadded dummy data collector"); } 

as see around end writing dummy data collection in end (10, 'something') , see hundreds of rows of data in _success file after job has terminated. can't identify why mapper code restarted multiple times on , on instead of running once. appreciated.

using jobconf.setnummaptasks(1) saying hadoop wish use 1 mapper, if possible, unlike setnumreducetasks, defines number specified.

that's why more mappers run , observe these numbers.

for more details, please read this post.