Upcoming rate limiting changes to Statuspage REST API

[#1]

Statuspage will soon be making changes to how rate limiting is implemented in our authenticated REST API. If you have an existing integration with our REST API, you may need to make updates to accommodate these changes, so please read on to understand the details.

What is changing

1. Rate limit policy changes. Effective Sep 15, 2020, unpaginated GET requests to the page-access-users and components endpoints will be limited to 1 request per minute. Paginated requests to those endpoints will continue to be rate limited at the existing rate of 60 requests per minute.
2. Rate limit response code change from 420 to 429. Effective Feb 15, 2021 when you exceed the Statuspage API rate limit, we’ll respond with HTTP status code 429 (Too Many Requests) instead of the current behavior of responding with a 420.
3. Inclusion of Retry-After header in rate limit responses. Effective immediately, we’ll include a Retry-After HTTP header in 420 / 429 responses, e.g. Retry-After: 60 for unpaginated requests that exceed the rate limit.

What you need to do

You may need to update your integration to prepare for these changes:

• We strongly recommend that you implement pagination for GET requests where possible, since the unpaginated rate limit will be much more restrictive.
• If you've implemented rate-limiting handling that expects a 420 status code, you should update it to also accommodate a 429 response.
• You should look for the Retry-After header and limit your requests accordingly.

If you have questions about these changes, please feel free to post a comment here, or contact our support team.

[Eric Taba]

Unfortunately, due to existing customer integrations built around the 420 response, we did not end up swapping the 420 response for 429 in the existing API v1 service. So in the case of rate-limiting, the response may be an HTTP 420 or a HTTP 429 depending on the service.

[Tarjei Utnes]

Hi!

Is there a rate limit on the non authenticated apis? Or is there some latency charts available for the open APIs?