Export from SQL MI with 100k records

[RazenC]

I have a Power App dashboard that has MS SQL connector and pagination feature. We have an export button that:

1. Calls a FLOW that passes the values from the filters from the app into the flow.

2. Calls an Azure Data Factory Pipeline that runs a stored procedure that creates a CSV file in blob

3. Copy the file from the blob and put it in the user's  One drive

with this approach we are able to get all the records we need. The issue is when we are now sharing the CANVAS APP to our Client they NEED TO HAVE ACCESS IN DATA FACTORY in order for that export to run. We definitely can't do that on security purposes.

Is there any alternate way to get this 100k+ records without Data factory doing the work?

Or maybe a way to use our login credentials in the flow for the connectors when other people are using the FLOW?

I saw some videos and links in youtube but this is more in power apps question regarding collections which only holds 2000 data each then pass it into the flow for processing in which it may work but the processing time is very very long. 

I also found this link in which I'm trying to copy but there are no response regarding this post so I'm not sure if it will work.

Exporting-SQL-Server-on-prem-table-with-over-300K-rows-to-CSV 

Thanks in advance.

[eskay · Accepted solution]

I think what you are searching for is the "Run Only Users" Function, found here:

In this menu, you can chose whether each person needs to supply its own connection to the (in your case) database or whether there should be a static one used.

[RazenC]

Great! I changed the connection to my EMAIL ADD on some connectors. I'm still testing this if this is gonna work for them when they ran the app then click the button that runs a flow. I'll let you know later or tomorrow if this works. Thanks!

[eskay]

Great to hear.

Side Note: I would propose to create a dedicated account for that connection, because then you could give that account only the needed rights, thus not running any risks regarding accidental or malicious writes and it would be easier to track, who did what in the database.

[RazenC]

Thanks for the suggestion. Will definitely raise this concern. As for the actual issue itself, yes it works. Thanks a lot!