Our APIs are protected with rate limiting. You are responsible for architecting and coding your application responsibly so as to not create too much traffic.

The token generated by this Portal has special rate limit restrictions and is not suitable for production use. Follow the Authentication guide to experience realistic rate limit restrictions. Note, however, that the current default rate limit for a given client ID is 50 requests per minute.

anchorResponse Format

Every rate limited request will have a 429 response with a retry-after header included. This means your request has not been processed. Do not try again until after the retry-after duration (in seconds) has passed, else you will receive another 429 response. For example, a 4 second wait time will look like so:

Retry-After: 4

anchorBest Practices

Before contacting Cisco for any rate limit issues, make sure you are following API best practices:

Architectural

If you are polling an API to find changes, consider subscribing to webhooks instead.

If you are polling on a regular interval (hourly, every 15m, etc.), consider scheduling different organizations or jobs at different times in the interval, perhaps using a hash.

Consider putting your own limits to prevent 1 customer from singlehandedly causing your app to hit rate limits.

Consider aggregating requests and using batch or bulk APIs.

For paginated APIs, avoid requesting unnecessary pages.

Coding

If user-driven, make sure to give the user feedback to try again shortly. You might consider showing the retry-after header value, too.

If the request is done in the background, wait for the retry-after header's period of time before trying again.

Make sure to limit the number of retries, to avoid infinite loops.