Brenno Alencar - Dados

Thanks for the help!

I use this approach of end_of_stream value and the cursor based incremental exports showed the same results. 

In another test, I use the Zenpy framework and in this case the export is ok. I guess Zenpy uses the API of query, but a didn't understand the difference in results in these API's.

I use this method: GET /api/v2/incremental/tickets.json?start_time={unix_time}

But on the return from the api I have fewer tickets than there actually are in zendesk. Any solution?