2048 row For Each loop limit - options for increase ?

(0) Share

Report

Posted on by DomAinSA

Hi There

I need to use a for each loop to cycle through a get rows query, and process each row to an HTTP endpoint. I have tried to build the process in Power Automate and in Azure Logic Apps, but in both processes, I am hitting a limit of 2048 records per loop. What would be the best way to solve for loops that need to process approx. 10000 rows per loop ?

Thanks

Dom

Categories:

Building flows

2025-07-04_155524...

I have the same question (0)

All responses (5)

Answers (0)

Ellis Karim 11,691 Super User 2025 Season 2 on at

Like (0)

Report
Could you explain a bit more about the overall process?

It would help to know:

What kind of data you're working with and where it’s coming from

Why each row needs to be sent individually to the HTTP endpoint

What the endpoint does with the data

Understanding the full scenario may help us figure out if there's a way to reduce the number of loops, and optimise the overall flow.

Ellis Karim
Blog | LinkedIn | Bluesky
If this solved your issue, please mark it as ✅ Accepted Answer. If it helped, feel free to give it a 🩷 Like!

Was this reply helpful? Yes No
Suggested answer

Infamous_Let 330 on at

Like (2)

Report

This is a common challenge with Power Automate and Logic Apps both have limits on how many items you can loop through at once. Since you need to process around 10,000 rows, you'll need to break it up a bit.

The most straightforward way is to enable pagination on your "Get Rows" step and then split the data into smaller chunks using the chunk() function. For example, you can break the full list into batches of 1000 and loop over each batch. This helps avoid hitting the loop limits and keeps things manageable.

If you're pulling from a database like SQL Server, a better option is to paginate directly in your SQL query using OFFSET and FETCH. That way, you only grab 1000 rows at a time and loop until there’s no more data.

For more advanced or large-scale processing, consider using Azure Durable Functions or Azure Data Factory. They’re built to handle bigger volumes and complex flows more efficiently than Logic Apps or Power Automate.

Was this reply helpful? Yes No
Chriddle 8,443 Super User 2025 Season 2 on at

Like (1)

Report

There's no such limit in Power Automate's "Apply to each"!
There may be a limit, but it's certainly higher than 2048 iterations. (Certainly higher than I'd like to try.)

Was this reply helpful? Yes No
DomAinSA 15 on at

Like (0)

Report

@Infamous_Let

This sounds like the approach that I need - do you have a visual example of how something like this would work ? I am struggling to picture it

Infamous_Let 190 on 04 Jul 2025 at 17:14:10

2048 row For Each loop limit - options for increase ?

This is a common challenge with Power Automate and Logic Apps both have limits on how many items you can loop through at once. Since you need to process around 10,000 rows, you'll need to break it up a bit.

The most straightforward way is to enable pagination on your "Get Rows" step and then split the data into smaller chunks using the chunk() function. For example, you can break the full list into batches of 1000 and loop over each batch. This helps avoid hitting the loop limits and keeps things manageable.

If you're pulling from a database like SQL Server, a better option is to paginate directly in your SQL query using OFFSET and FETCH. That way, you only grab 1000 rows at a time and loop until there’s no more data.

For more advanced or large-scale processing, consider using Azure Durable Functions or Azure Data Factory. They’re built to handle bigger volumes and complex flows more efficiently than Logic Apps or Power Automate.

Was this reply helpful? Yes No
Suggested answer

Infamous_Let 330 on at

Like (1)

Report

I believe if you build this flow it will help you understand how it works

in the Get Items go to settings and turn on pagination and set the threshold higher than the expected amount of records as shown below

Was this reply helpful? Yes No