After upgrading Splunk DB Connect, why am I getting error "invalid literal for float()" when I create a db input?

[dbrancaglion]

I can see the results without problem in the query page, but if I create a db input, Splunk shows me the following error in /opt/splunk/var/log/splunk/dbx2.log :

[ERROR] [modular_input_event_writer.py], line 102: action=converting_incoming_data_in_csv_to_event_stream_failed input_mode=batch dbinput="mi_input://rec_limit_exporter" error="invalid literal for float(): 1471289753,172"

This error happened after I upgraded my Splunk DB Connect. Can I roll this back?

I tried to reinstall all Splunk DB Connect 2, but with no success.

Someone have a similar problem?

[dlisicre]

As indicated in mhornste thread, changing output_timestamp_format = dd-MM-yyyy HH:mm:ss to Epoch managed to solve this issue for me.
Kind regards,
David Lisin

[dlisicre]

Same issue here at the moment.

[mhornste]

I have the same problem. I'm not using SSL and enable_query_wrapping = 0/ 1 don't work (both of them).

[jcoates_splunk]

Try this. http://docs.splunk.com/Documentation/DBX/2.3.0/DeployDBX/Troubleshooting#Database_inputs_or_lookups_...