Big Data: Sqoop-Export Error

Question

I am very new to this world. While running export command using sqoop, I am getting the below error “Input path does not exist: hdfs://quickstart.cloudera:8020/home/cloudera/Test5”. I have checked the path /home/cloudera/Test5 and the file exists in the path. From the core-site.xml file of sqoop configuration the details of hdfs path is coming, when I tested it through file browser Just opening IE and type hdfs://quickstart.cloudera:8020/home/cloudera/Test5, the message is coming as “Unable to connect”. I do not know the correct paramater values of the property. Please help me in solving this issue.

Please find the property file parameter and errir details below.

Parameter file

<name>fs.defaultFS</name>
<value>hdfs://quickstart.cloudera:8020</value>

Error

[cloudera@quickstart hadoop-conf]$ sqoop export --connect jdbc:sqlserver://10.34.83.177:54815 --username hadoop --password xxxxxx --table hadoop_sanjeeb3 --export-dir /home/cloudera/Test5 -mapreduce-job-name sqoop_export_job -m 1
Warning: /usr/lib/sqoop/../accumulo does not exist! Accumulo imports will fail.
Please set $ACCUMULO_HOME to the root of your Accumulo installation.
15/10/01 08:42:47 INFO sqoop.Sqoop: Running Sqoop version: 1.4.5-cdh5.4.2
15/10/01 08:42:47 WARN tool.BaseSqoopTool: Setting your password on the command-line is insecure. Consider using -P instead.
15/10/01 08:42:48 INFO manager.SqlManager: Using default fetchSize of 1000
15/10/01 08:42:48 INFO tool.CodeGenTool: Beginning code generation
15/10/01 08:42:49 INFO manager.SqlManager: Executing SQL statement: SELECT t.* FROM [hadoop_sanjeeb3] AS t WHERE 1=0
15/10/01 08:42:49 INFO orm.CompilationManager: HADOOP_MAPRED_HOME is /usr/lib/hadoop-mapreduce
Note: /tmp/sqoop-cloudera/compile/aa9c9fd9f69b76202be29508561f22ff/hadoop_sanjeeb3.java uses or overrides a deprecated API.
Note: Recompile with -Xlint:deprecation for details.
15/10/01 08:42:51 INFO orm.CompilationManager: Writing jar file: /tmp/sqoop-cloudera/compile/aa9c9fd9f69b76202be29508561f22ff/hadoop_sanjeeb3.jar
15/10/01 08:42:51 INFO mapreduce.ExportJobBase: Beginning export of hadoop_sanjeeb3
15/10/01 08:42:51 INFO Configuration.deprecation: mapred.job.tracker is deprecated. Instead, use mapreduce.jobtracker.address
15/10/01 08:42:51 INFO Configuration.deprecation: mapred.jar is deprecated. Instead, use mapreduce.job.jar
15/10/01 08:42:54 WARN mapreduce.ExportJobBase: Input path hdfs://quickstart.cloudera:8020/home/cloudera/Test5 does not exist
15/10/01 08:42:54 INFO Configuration.deprecation: mapred.reduce.tasks.speculative.execution is deprecated. Instead, use mapreduce.reduce.speculative
15/10/01 08:42:54 INFO Configuration.deprecation: mapred.map.tasks.speculative.execution is deprecated. Instead, use mapreduce.map.speculative
15/10/01 08:42:54 INFO Configuration.deprecation: mapred.map.tasks is deprecated. Instead, use mapreduce.job.maps
15/10/01 08:42:54 INFO client.RMProxy: Connecting to ResourceManager at /0.0.0.0:8032
15/10/01 08:42:58 INFO mapreduce.JobSubmitter: Cleaning up the staging area /tmp/hadoop-yarn/staging/cloudera/.staging/job_1443557935828_0011
15/10/01 08:42:58 WARN security.UserGroupInformation: PriviledgedActionException as:cloudera (auth:SIMPLE) cause:org.apache.hadoop.mapreduce.lib.input.InvalidInputException: Input path does not exist: hdfs://quickstart.cloudera:8020/home/cloudera/Test5
15/10/01 08:42:58 ERROR tool.ExportTool: Encountered IOException running export job: org.apache.hadoop.mapreduce.lib.input.InvalidInputException: Input path does not exist: hdfs://quickstart.cloudera:8020/home/cloudera/Test5

Regards - Sanjeeb

Answer 1

That file may be in the local file system, but not in the hadoop distributed file system (HDFS). You can add those local files from local file system to HDFS by

  hadoop fs -put <local_file_path> <HDFS_diresctory> 

command. You should do it as an HDFS user.

Answer 2

Seems like you have some confusion between local file system and hadoop file system. The file that you are trying to export using sqoop should be present in hdfs. The directory location /home/cloudera/Test seems to be present in local file system.

Execute the command given below and confirm that the location that you mentioned exists in hdfs.

hadoop fs -ls /home/cloudera/Test5

If this is giving error, it means the directory doesn't exists in hdfs. You can't browse hdfs with a simple ls command, you have to use hadoop commands. If you want to browse hdfs directories using a browser, open the namenode web ui (http://namenode-host:50070) and there you have an option to browse the files and directories. You cannot browse the hdfs filesystem using a url like hdfs://quickstart.cloudera:8020/home/cloudera/Test5 using the browser. You can use webhdfs for similar operation.

Ensure that the file is present in hdfs and trigger the command again. It will work.

NB: Usually we never keep user directories like /home/cloudera in hdfs. The structure will be something like /user/{username}. By default, hdfs considers /user/{username} as the home dir in hdfs. Where {username} will be the current logged-in user in linux