Athena really isn't designed for single record inserts as each insert will
create another file in S3 and the driver behaves a lot more like Hive than a
regular jdbc connection so that processor probably won't ever work. To load
data into Athena from NiFi you can either use ConvertRecord to convert to
Parquet and then put the files directly into S3 or use the PutParquet processor
after adding the S3 dependencies to the class path for the processor. This
archive messaged explains what you need to do to do that
http://apache-nifi-users-list.2361937.n4.nabble.com/PutParquet-with-S3-td3632.html.
Thanks
Shawn
On 1/8/20, 12:33 PM, "apacheco" <anton...@seatfrog.com> wrote:
Hi,
I am relatively new to Nifi, and am trying to use PutDatabaseRecord using
the Athena JDBC 42 driver to populate an AWS Athena Table. I have designed
the flow to generate JSON records out of a large XML file, and in order to
debug the flow I have split the flow into to two PutDatabaseRecord
processors, one for a MySQL table, and the other to the Athena one. There is
no issues using an INSERT statement into MySQL, but with Athena I am getting
a "*[Simba][JDBC](10940) Invalid parameter index: 1*".
Is *INSERT INTO* using JDBC with PutDatabaseRecord not supported for AWS
Athena?
<http://apache-nifi-users-list.2361937.n4.nabble.com/file/t837/a.png>
--
Sent from: http://apache-nifi-users-list.2361937.n4.nabble.com/
