0
    [Hive side]
    create table TEST.HIVE_TEST (
    CONTENT string)
    stored as textfile;
    
    [oracle side]
    create table TEST.ORACLE_TEST (
    CONTENT clob);
    
    [sqoop export]
    /usr/bin/sqoop export \
    --connect "connection string"
    --username "ex" password "ex" \
    --table ORACLE_TEST \
    --columns CONTENT \
    --hcatalog-table HIVE_TEST --hcatalog-database TEST \
    --map-column-java CONTENT=String

result of mapreduce is successed but when i see the yarn application log i find ERROR [main] org.apache.sqoop.mapreduce.hcat.SqoopHCatExportHelper : cannot convert hcatalog object of type string to java object type org.apache.sqoop.lib.clobref

and i see null in oracle what is the solution can anyone help?

sqoop version is Sqoop-1.4.7-cdh6.3.4

i hope anyone knows how to export data from string to clob(oracle) using sqoop

are123
  • 1
  • 1
  • You can follow this solution. You need to change case or oracle clob column to lowercase and try your sqoop export command https://issues.apache.org/jira/browse/SQOOP-3254 – Koushik Roy Dec 08 '22 at 09:55
  • when i created table like the DDL in the link you gave me. but when i sqoop export, the error message is like "ORA-00904: "CLOB_COLUMN" invalid identifer – are123 Dec 09 '22 at 07:30
  • they used `clob col` name, tiyre using `content`. you need to replace using your colname – Koushik Roy Dec 09 '22 at 09:18
  • of course, i changed the export command from uppsercase to lowercase but still same error.. – are123 Dec 12 '22 at 01:34

0 Answers0