N8n Coda API "Get All Rows" Stuck at 498/500 Items

1

Hello Coda Community,

I’m working on extracting a large number of rows from a Coda table using n8n and pushing them to another service. My Coda table currently contains over 5000 rows.

However, I’m consistently only able to retrieve approximately 498-500 items per execution, even though my table clearly has many more rows.

Here’s what I’ve tried to resolve this:

  1. Using the native n8n Coda Node:
  • Operation: Get All Rows for a specific table.
  • The “Return All” toggle within the node is ON, which is intended to handle pagination automatically.
  • Result: The node only outputs ~500 items.
  1. Using an n8n HTTP Request Node (Manual Pagination):
  • Endpoint Structure: https://coda.io/apis/v1/docs/{docId}/tables/{tableIdOrName}/rows
  • Method: GET
  • Authentication: Bearer token via Authorization header. I’ve confirmed the API key is valid and has read permissions for the document/table (and have generated/re-checked keys multiple times to rule out credential issues).
  • Query Parameters:
    • limit: Set to 500 (the Coda API’s max per page).
    • pageToken: Dynamically passed using an n8n expression ({{ $json.nextPageToken }}) to retrieve subsequent pages, with “Keep Empty Fields” enabled for the initial request.
  • Workflow Logic: Implemented a Do While loop in n8n to continuously fetch pages using the nextPageToken from each response until no more tokens are returned.
  • Result: Despite the loop and pageToken parameter, the process still seems to hit a ceiling at around 498-500 items, and a nextPageToken for further data does not appear in the API response, even though many more rows exist in the actual Coda table.

My questions for the community are:

  • Is there any specific behavior of the Coda API’s rows endpoint that might prevent the nextPageToken from being returned after the first page (or first ~500 items) when dealing with large datasets?
  • Are there any less common API parameters or best practices for handling pagination for very large tables in Coda that I might be missing?
  • Could there be a subtle Coda-side configuration or data state that could influence how the API reports total pages or provides nextPageToken?

Any insights or troubleshooting tips would be greatly appreciated!

Thank you!

1 Like
2

Hi @Raj_Rathod ! Your logic seems to be correct, when working with n8n I’ve had more success using HTTP nodes for the API calls than the native Coda nodes.

Are you setting a batch limit in your http node?

3

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.