| | --- |
| | |
| | title: Handling API rate limits |
| | contentType: howto |
| | description: How to handle API rate limits when using n8n integrations. |
| | --- |
| | |
| | # Handling API rate limits |
| |
|
| | [API](/glossary.md#api) rate limits are restrictions on request frequency. For example, an API may limit the number of requests you can make per minute, or per day. |
| |
|
| | APIs can also limits how much data you can send in one request, or how much data the API sends in a single response. |
| |
|
| | ## Identify rate limit issues |
| |
|
| | When an n8n node hits a rate limit, it errors. n8n displays the error message in the node output panel. This includes the error message from the service. |
| |
|
| | If n8n received error 429 (too many requests) from the service, the error message is **The service is receiving too many requests from you**. |
| |
|
| | To check the rate limits for the service you're using, refer to the API documentation for the service. |
| |
|
| | ## Handle rate limits for integrations |
| |
|
| | There are two ways to handle rate limits in n8n's integrations: using the Retry On Fail setting, or using a combination of the [Loop Over Items](/integrations/builtin/core-nodes/n8n-nodes-base.splitinbatches.md) and [Wait](/integrations/builtin/core-nodes/n8n-nodes-base.wait.md) nodes: |
| |
|
| | * Retry On Fail adds a pause between API request attempts. |
| | * With Loop Over Items and Wait you can break you request data into smaller chunks, as well as pausing between requests. |
| |
|
| | ### Enable Retry On Fail |
| |
|
| | When you enable Retry On Fail, the node automatically tries the request again if it fails the first time. |
| |
|
| | 1. Open the node. |
| | 1. Select **Settings**. |
| | 1. Enable the **Retry On Fail** toggle. |
| | 1. Configure the retry settings: if using this to work around rate limits, set **Wait Between Tries (ms)** to more than the rate limit. For example, if the API you're using allows one request per second, set **Wait Between Tries (ms)** to `1000` to allow a 1 second wait. |
| |
|
| | ### Use Loop Over Items and Wait |
| |
|
| | Use the Loop Over Items node to batch the input items, and the Wait node to introduce a pause between each request. |
| |
|
| | 1. Add the Loop Over Items node before the node that calls the API. Refer to [Loop Over Items](/integrations/builtin/core-nodes/n8n-nodes-base.splitinbatches.md) for information on how to configure the node. |
| | 1. Add the Wait node after the node that calls the API, and connect it back to the Loop Over Items node. Refer to [Wait](/integrations/builtin/core-nodes/n8n-nodes-base.wait.md) for information on how to configure the node. |
| |
|
| | For example, to handle rate limits when using OpenAI: |
| |
|
| |  |
| |
|
| | ## Handle rate limits in the HTTP Request node |
| |
|
| | The HTTP Request node has built-in settings for handling rate limits and large amounts of data. |
| |
|
| | ### Batch requests |
| |
|
| | Use the Batching option to send more than one request, reducing the request size, and introducing a pause between requests. This is the equivalent of using Loop Over Items and Wait. |
| |
|
| | 1. In the HTTP Request node, select **Add Option** > **Batching**. |
| | 1. Set **Items per Batch**: this is the number of input items to include in each request. |
| | 1. Set **Batch Interval (ms)** to introduce a delay between requests. For example, if the API you're using allows one request per second, set **Wait Between Tries (ms)** to `1000` to allow a 1 second wait. |
| |
|
| | ### Paginate results |
| |
|
| | APIs paginate their results when they need to send more data than they can handle in a single response. For more information on pagination in the HTTP Request node, refer to [HTTP Request node | Pagination](/integrations/builtin/core-nodes/n8n-nodes-base.httprequest/index.md#pagination). |
| |
|
