borismo commented on issue #25904:
URL: https://github.com/apache/airflow/issues/25904#issuecomment-2671074409

   > As mentioned before, we can support connection methods that Tableau SDK 
allows. If the SDK doesn't offer the needed option then I suggest to contact 
Tableau and ask them to address it.
   
   In the [official doc of the Python 
SDK](https://tableau.github.io/server-client-python/docs/sign-in-out), I read:
   > Sign in with Personal Access Token
   In most cases this is the preferred method because it improves security by 
avoiding the need to use or store passwords directly.
   
   This is the SDK used by the provider, right?
   
   In our team we have been using the provider with the PAT successfully, aware 
of the limitation of one active connection at a time. This is how we mitigate 
it when refreshing data sources (our main use case of the provider):
   - one pool of 1 slot per connection
   - set `blocking_refresh=False` on `TableauOperator`
   - use `TableauJobStatusSensor` to wait for the refresh to complete
   
   That way , we can refresh multiple data sources concurrently with a single 
connection. I hope it's useful to anyone. Having multiple connection is also a 
valid option.
   
   With the removal of PAT support in version 5.0.0, as Tableau Cloud users 
with MFA enforced, as far as I understand, we can't use this provider anymore. 
Maybe other teams are impacted similarly?
   
   So I'm having a hard time understanding why support for PAT was deprecated 
and dropped. Am I missing something obvious?
   If not, shall it be brought back, with a section in the docs to explain how 
to work around PAT's limitation? (happy to contribute to the doc).


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to