Rate limit
also: rate limits, rate limiting, rate-limited, rate limited, throttled, throttling
A cap on how many requests you can send to a service in a set time, like 100 per minute. Go over it, and you're refused until time passes.
A policy restricting request volume per client, key or IP address over a time window. Going over usually returns HTTP 429 Too Many Requests, often with a Retry-After header. You can hit other services' limits, and set your own to protect your API from abuse.
A cafe's "two free refills" rule. Ask for a third and the barista says no. The coffee machine isn't broken; you've just hit the limit.
Rate limits explain many "worked in testing, fails at scale" bugs. Setting your own protects sign-up and login forms from bots. Agents bring them up when imports or bulk jobs start failing.
The sync is hitting GitHub's rate limit. I'll slow down and retry after a pause, and save results so we make fewer calls.
Add a rate limit to the login and password-reset endpoints so a single IP address, meaning one network connection, can't try more than 10 times a minute.
Assuming the service is down or your code is broken. A 429 means you're sending too much, too fast. Slow down and retry later.