SQOOP issues with ORACLE datatypes - RAW, BFILE, LONG RAW

[#1]

Example oracle RAW datatype value “175D78D86FAFE6D19C631AF3BFC246EB” and the sqoop converts that data into string. The data has spaces when I check the hdfs file using "hive –orcdump –d <hdfsfile>”. The RAW dataype is expressed in hexadecimal on oracle side and sqoop adds space to the hdfs file like “17 5D 78 D8 6F AF E6 D1 9C 63 1A F3 BF C2 46 EB”. Has anyone faced this ?

RAW datatype is supported by sqoop in the documentation - https://sqoop.apache.org/docs/1.4.6/SqoopUserGuide.html

I see BFILE and LONG RAW datatypes are not supported in sqoop documentation. Do we have a workaround for this ?

[#2 · Accepted solution]

@nbalaji-elangovan

Have you looked into Type Mapping to override the data types?

https://sqoop.apache.org/docs/1.4.6/SqoopUserGuide.html#_controlling_type_mapping

[#3]

Ya I tried it and that doesn't help.

[#4]

This has been accepted as a bug by the support team and tracked in the internal JIRA.

[#5]

Hi balaji what is the status of this Jira. did you have any referance to that jira I can refer it ?