@EricL    I tried -map-column-java Settings=String and got the data type as string in hive. however the data doesnt look corrupted but the hexadecimal representation of the value is not retained   cmd used :  sqoop import --driver 'com.microsoft.sqlserver.jdbc.SQLServerDriver' --connect <> --connection-manager 'org.apache.sqoop.manager.SQLServerManager' --username <> -password <> --table 'my_table' --as-parquetfile --delete-target-dir --target-dir /user/test/ -map-column-java Settings=String --m 1    Result of above query:  column data type in source : image column value in source : 0xFFffDeDJBF.......dDf.      --> ( A hexadecimal value ) column data type in hive: string  column value in hive: FFffDeDJBF.......dDf (0x is not retained)         ... View more

@EricL    I tried the above. the map-column-hive is not making any impact. the column data type is taken as string in hive after import and the data is still corrupted.   Using hive-import and hive-table with it helped in getting the data type as binary in hive. But the data is still corrupted cmd used : sqoop import --driver 'com.microsoft.sqlserver.jdbc.SQLServerDriver' --connect <> --connection-manager 'org.apache.sqoop.manager.SQLServerManager' --username <> -password <> --hive-import --hive-table testc --as-parquetfile --delete-target-dir --target-dir '/user/test/' --query "select * FROM my_table where \$CONDITIONS; " --m 1 --map-column-hive Settings=binary   ... View more

@VidyaSargur    Any sqoop expert who can throw some light on this issue ?   ... View more

@pmohan    Does this help with columns having data type - image ??   converting the image data type to binary again is leading to data corruption. Please see the issue highlighted below :  https://community.cloudera.com/t5/Support-Questions/SQOOP-import-of-quot-image-quot-data-type-into-hive/td-p/283584 ... View more

@Khanna      command used : sqoop import --driver "com.microsoft.sqlserver.jdbc.SQLServerDriver" --connect "jdbc:sqlserver://server:port;database=db_name;" --connection-manager "org.apache.sqoop.manager.SQLServerManager" --username <> -password <> --table 'table_name' --as-parquetfile --delete-target-dir --target-dir /user/test/axe/ --temporary-rootdir /user/test2/ --m 4 --split-by user_id   before mapreduce starts: hadoop fs -ls -R /user/test/ /user/test/axe /user/test/axe/.metadata /user/test/axe/.metadata/descriptor.properties /user/test/axe/.metadata/schema.avsc /user/test/axe/.metadata/schemas /user/test/axe/.metadata/schemas/1.avsc   when the mapreduce job starts: hadoop fs -ls -R /user/test/ /user/test/.temp /user/test/.temp/job_1571067970221_0156 /user/test/.temp/job_1571067970221_0156/mr /user/test/.temp/job_1571067970221_0156/mr/job_1571067970221_0156 /user/test/.temp/job_1571067970221_0156/mr/job_1571067970221_0156/.metadata /user/test/.temp/job_1571067970221_0156/mr/job_1571067970221_0156/.metadata/descriptor.properties /user/test/.temp/job_1571067970221_0156/mr/job_1571067970221_0156/.metadata/schema.avsc /user/test/.temp/job_1571067970221_0156/mr/job_1571067970221_0156/.metadata/schemas /user/test/.temp/job_1571067970221_0156/mr/job_1571067970221_0156/.metadata/schemas/1.avsc /user/test/axe /user/test/axe/.metadata /user/test/axe/.metadata/descriptor.properties /user/test/axe/.metadata/schema.avsc /user/test/axe/.metadata/schemas /user/test/axe/.metadata/schemas/1.avsc   once the import is complete: hadoop fs -ls -R /user/test/ /user/test/axe /user/test/axe/.metadata /user/test/axe/.metadata/descriptor.properties /user/test/axe/.metadata/schema.avsc /user/test/axe/.metadata/schemas /user/test/axe/.metadata/schemas/1.avsc /user/test/axe/.signals /user/test/axe/.signals/unbounded /user/test/axe/679dadfc-3657-410b-a805-6c98b8d1720b.parquet /user/test/axe/923bbf06-35d6-4156-8df3-a6e53ebf00f7.parquet /user/test/axe/aef24eac-15dd-4ebc-a61a-68265d53a320.parquet /user/test/axe/f903a079-87b3-48a0-bea6-aa02d92c5aac.parquet   ... View more

@Khanna    I tried it. but the .temp is still getting created at the target directory that i specify.     ... View more

@Shelton    The target-dir is specified so that the generated files are placed in that directory. The problem I am facing here is with the temporary directory (.temp) that gets created during run time (i.e when the mapreduce job is initiated) at the target -dir. The solution I am looking for is to change the temporary directory(.temp) location. ... View more