Sqoop export error - cause:org.apache.hadoop.mapre

Question

I am developing a java program.

The java program exports data from hive to mysql.

First, I write the code

ProcessBuilder pb = new ProcessBuilder("sqoop-export", "export", 
         "--connect",               "jdbc:mysql://localhost/mydb", 
         "--hadoop-home",    "/home/yoonhok/development/hadoop-1.1.1", 
         "--table",                    "mytable", 
         "--export-dir",            "/user/hive/warehouse/tbl_2", 
         "--username",            "yoonhok", 
         "--password",            "1234");

try {
    Process p = pb.start();
    if (p.waitFor() != 0) {
        System.out.println("Error: sqoop-export failed.");
        return false;
    }
} catch (IOException e) {
    e.printStackTrace();
} catch (InterruptedException e) {
    e.printStackTrace();
}

It works perfectly.

But I learned a new way of using sqoop in java.

Sqoop doesn't support client api yet.

So I added sqoop lib and just write Sqoop.run()

Second, I write the code again with new way.

String[] str = {"export", 
     "--connect",               "jdbc:mysql://localhost/mydb", 
     "--hadoop-home",    "/home/yoonhok/development/hadoop-1.1.1", 
     "--table",                    "mytable", 
     "--export-dir",            "/user/hive/warehouse/tbl_2", 
     "--username",            "yoonhok", 
     "--password",            "1234"
};

if (Sqoop.runTool(str) == 1) {
     System.out.println("Error: sqoop-export failed.");
     return false;
}

But It doesn't running.

I got error......

13/02/14 16:17:09 WARN tool.BaseSqoopTool: Setting your password on the command-line is insecure. Consider using -P instead. 
13/02/14 17:43:12 WARN sqoop.ConnFactory: $SQOOP_CONF_DIR has not been set in the environment. Cannot check for additional configuration.
13/02/14 16:17:09 INFO manager.MySQLManager: Preparing to use a MySQL streaming resultset. 
13/02/14 16:17:09 INFO tool.CodeGenTool: Beginning code generation 
13/02/14 16:17:09 INFO manager.SqlManager: Executing SQL statement: SELECT t.* FROM `tbl_2` AS t LIMIT 1 
13/02/14 16:17:09 INFO manager.SqlManager: Executing SQL statement: SELECT t.* FROM `tbl_2` AS t LIMIT 1 
13/02/14 16:17:09 INFO orm.CompilationManager: HADOOP_HOME is /home/yoonhok/development/hadoop-1.1.1 
Note: /tmp/sqoop-yoonhok/compile/45dd1a113123726796a4ed4ce10c9110/tbl_2.java uses or overrides a deprecated API. 
Note: Recompile with -Xlint:deprecation for details. 
13/02/14 16:17:10 INFO orm.CompilationManager: Writing jar file: /tmp/sqoop-yoonhok/compile/45dd1a113123726796a4ed4ce10c9110/tbl_2.jar 
13/02/14 16:17:10 INFO mapreduce.ExportJobBase: Beginning export of tbl_2 
13/02/14 16:17:10 WARN mapreduce.ExportJobBase: Input path file:/user/hive/warehouse/tbl_2 does not exist 
13/02/14 16:17:11 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable 
13/02/14 16:17:11 INFO mapred.JobClient: Cleaning up the staging area file:/tmp/hadoop-yoonhok/mapred/staging/yoonhok314601126/.staging/job_local_0001 
13/02/14 16:17:11 ERROR security.UserGroupInformation: PriviledgedActionException as:yoonhok cause:org.apache.hadoop.mapreduce.lib.input.InvalidInputException: Input path does not exist: file:/user/hive/warehouse/tbl_2 
13/02/14 16:17:11 ERROR tool.ExportTool: Encountered IOException running export job: org.apache.hadoop.mapreduce.lib.input.InvalidInputException: Input path does not exist: file:/user/hive/warehouse/tbl_2

I saw $SQOOP_CONF_DIR has not been set in the environment.

so I added

SQOOP_CONF_DIR=/home/yoonhok/development/sqoop-1.4.2.bin__hadoop-1.0.0/conf

in the

/etc/environment

And try again, but Error...

13/02/14 16:17:09 WARN tool.BaseSqoopTool: Setting your password on the command-line is insecure. Consider using -P instead. 
13/02/14 16:17:09 INFO manager.MySQLManager: Preparing to use a MySQL streaming resultset. 
13/02/14 16:17:09 INFO tool.CodeGenTool: Beginning code generation 
13/02/14 16:17:09 INFO manager.SqlManager: Executing SQL statement: SELECT t.* FROM `tbl_2` AS t LIMIT 1 
13/02/14 16:17:09 INFO manager.SqlManager: Executing SQL statement: SELECT t.* FROM `tbl_2` AS t LIMIT 1 
13/02/14 16:17:09 INFO orm.CompilationManager: HADOOP_HOME is /home/yoonhok/development/hadoop-1.1.1 
Note: /tmp/sqoop-yoonhok/compile/45dd1a113123726796a4ed4ce10c9110/tbl_2.java uses or overrides a deprecated API. 
Note: Recompile with -Xlint:deprecation for details. 
13/02/14 16:17:10 INFO orm.CompilationManager: Writing jar file: /tmp/sqoop-yoonhok/compile/45dd1a113123726796a4ed4ce10c9110/tbl_2.jar 
13/02/14 16:17:10 INFO mapreduce.ExportJobBase: Beginning export of tbl_2 
13/02/14 16:17:10 WARN mapreduce.ExportJobBase: Input path file:/user/hive/warehouse/tbl_2 does not exist 
13/02/14 16:17:11 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable 
13/02/14 16:17:11 INFO mapred.JobClient: Cleaning up the staging area file:/tmp/hadoop-yoonhok/mapred/staging/yoonhok314601126/.staging/job_local_0001 
13/02/14 16:17:11 ERROR security.UserGroupInformation: PriviledgedActionException as:yoonhok cause:org.apache.hadoop.mapreduce.lib.input.InvalidInputException: Input path does not exist: file:/user/hive/warehouse/tbl_2 
13/02/14 16:17:11 ERROR tool.ExportTool: Encountered IOException running export job: org.apache.hadoop.mapreduce.lib.input.InvalidInputException: Input path does not exist: file:/user/hive/warehouse/tbl_2

I think that Export-dir is problem.

I use "/user/hive/warehouse/tbl_2".

And When I run "hadoop fs -ls /user/hive/warehouse/", the table "tbl_2" exist.

I think that

"Input path does not exist: file:/user/hive/warehouse/tbl_2" is not ok.

"Input path does not exist: hdfs:/user/hive/warehouse/tbl_2" is ok.

But I don't know how can I fix it.

---

Ok just before I got a hint.

And I edited 'export-dir'

--export-dir   hdfs://localhost:9000/user/hive/warehouse/tbl_2

But... It's error... T.T

13/02/15 15:17:20 WARN tool.BaseSqoopTool: Setting your password on the command-line is insecure. Consider using -P instead.
13/02/15 15:17:20 INFO manager.MySQLManager: Preparing to use a MySQL streaming resultset.
13/02/15 15:17:20 INFO tool.CodeGenTool: Beginning code generation
13/02/15 15:17:20 INFO manager.SqlManager: Executing SQL statement: SELECT t.* FROM `tbl_2` AS t LIMIT 1
13/02/15 15:17:20 INFO manager.SqlManager: Executing SQL statement: SELECT t.* FROM `tbl_2` AS t LIMIT 1
13/02/15 15:17:20 INFO orm.CompilationManager: HADOOP_HOME is /home/yoonhok/development/hadoop-1.1.1/libexec/..
Note: /tmp/sqoop-yoonhok/compile/697590ee9b90c022fb8518b8a6f1d86b/tbl_2.java uses or overrides a deprecated API.
Note: Recompile with -Xlint:deprecation for details.
13/02/15 15:17:22 INFO orm.CompilationManager: Writing jar file: /tmp/sqoop-yoonhok/compile/697590ee9b90c022fb8518b8a6f1d86b/tbl_2.jar
13/02/15 15:17:22 INFO mapreduce.ExportJobBase: Beginning export of tbl_2
13/02/15 15:17:22 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
13/02/15 15:17:23 INFO input.FileInputFormat: Total input paths to process : 1
13/02/15 15:17:23 INFO input.FileInputFormat: Total input paths to process : 1
13/02/15 15:17:23 INFO mapred.JobClient: Cleaning up the staging area file:/tmp/hadoop-yoonhok/mapred/staging/yoonhok922915382/.staging/job_local_0001
13/02/15 15:17:23 ERROR security.UserGroupInformation: PriviledgedActionException as:yoonhok cause:java.io.FileNotFoundException: File /user/hive/warehouse/tbl_2/000000_0 does not exist.
13/02/15 15:17:23 ERROR tool.ExportTool: Encountered IOException running export job: java.io.FileNotFoundException: File /user/hive/warehouse/tbl_2/000000_0 does not exist.

When I checked hdfs,

hadoop fs -ls /user/hive/warehouse/tbl_2

or

hadoop fs -ls hdfs://localhost:9000/user/hive/warehouse/tbl_2

the file exist.

-rw-r--r-- 1 yoonhok supergroup 14029022 2013-02-15 12:16 /user/hive/warehouse/tbl_2/000000_0

I try in the shell command in terminal

sqoop-export --connect jdbc:mysql://localhost/detector --table tbl_2 --export-dir hdfs://localhost:9000/user/hive/warehouse/tbl_2 --username yoonhok --password 1234

It's work.

What's problem?

I don't know.

Could you help me?

Answer 1

You need to load and provide your Hadoop configuration files. By default they are read from classpath, but you might be able to override this by Configuration.setDefaultResource (without guarantees).