Skip to main content
Irene Strodel ๋‹˜์ด #Tableau Cloud์— ์งˆ๋ฌธํ–ˆ์Šต๋‹ˆ๋‹ค

I have created a data source in Tableau desktop combining several tables on Amazon Redshift and some Google sheets. Then I created an extract and published the data source to Tableau online, to use it in other workbooks.

At the end of April I added another dimension to some tables and updated to 2021.1.1 and republished the data source. The new diminution (campaign name) did not increase the level of detail, as I was using ad name (higher granularity than campaign name) already - see screenshots attached with the rows per table.

Since this time the scheduled extract refresh runs every morning and I don't get an error message, that the refresh failed. BUT every time the data is set back to April 26th.

A manual refresh on Tableau online has the same result - the only "solution" I have at the moment is to open Tableau Desktop, refresh the extract there and republish it.

Changing to a live connection is not an option, as then the connected workbook won't load.

Things I tried:

  • publishing the DS as live and creating the extract online
  • downgrading to 2020.4 and publishing the DS from this workbook

Is there a time-out on Tableau online for creating extracts - offline it can take up to 20 min to create the extract? But if, so, why don't I get an "extract creation failed" email?

I've contacted Tableau support last week, but never heard back - and cannot find the ticket in my emails or profile to follow up. ๐Ÿคทโ€โ™€๏ธ

Any recommendations how to solve this are welcome.

๋‹ต๋ณ€ 2๊ฐœ
  1. 2021๋…„ 6์›” 18์ผ ์˜ค์ „ 7:30

    Hi @Peter Fakanโ€‹ 

    thank you for your suggestion. Unfortunately, none of them solved the issue.

    I had to replace my primary google sheet, which was just a union of different sheets to create a row for every date-country combination. So, not a "big" google sheet with lots of data. But even when the sheet itself had dates until yesterday, Tableau struggled to keep them updated.

    So, I connected to a date-time table on Amazon redshift and multiplied the data per country using custom SQL.

    Hi @Peter Fakanโ€‹ thank you for your suggestion. Unfortunately, none of them solved the issue.It looks like this is working reliably for about a week now.

0/9000