Hello everyone,
We have encountered some issues lately while pushing data from HDFS to Teradata. We already found out that “–columns” doesn’t work during export.
Now we are facing another issue, not actually an issue may be. How we map between HDFS output to Teradata table? During, Oracle export we can use –columns and it works perfectly. But when exporting to Teradata, the same configuration doesn’t work.
Say, in my HDFS, I have a output file like this:
columnA columnB ColumnC columnD
In Teradata, if the table was created with a different order, say
columnB columnD ColumnC columnA
Then what our options to map the output to Teradata table? Or in current Hortonworks Sqoop Teradata connector, there is no option for that and the Teradata table needs to follow exactly the same order as output?
We have tried target.field.names and source.field.names options already. Any help would be highly appreciated.
Thank you.