To ensure the stability of our API, we have recently added extra rate limiting to our infrastructure. This includes the maximum number of calls per IP or per token.

There are two types of rate limiting on the REST API, both of which are explained below. Rate limiting is used to manage traffic from external systems and protect the Deployteq platform from sudden high loads on the REST API and Deployteq systems.

Type:

Number of connections per IP

Explanation: An IP address may establish a maximum of 20 connections per Deployteq proxy; the 21st connection will simply be rejected. Currently, there are four active Deployteq proxies, which means that each IP address can establish a maximum of (4 * 20) 80 connections at the same time. This rate limit is almost never reached but is in place for added safety.

Type:

Number of calls per token or IP address

Explanation: A token may execute a maximum of 25 calls simultaneously. This has a Burst applied to it, which means that if 25 is reached, you can continue to send calls. 100 requests in addition to the 25 are accepted but processed at a maximum rate of 25 at a time. If you exceed 100 calls that are in the queue, you will receive a 429 error. In addition to the 100, 400 may be submitted at the same time and will be accepted but processed at a maximum rate of 100 at a time.