Rate limiting

The API accepts a maximum number of requests per time window. This limit — rate limiting — protects the platform from excess, such as an integration stuck in a loop that fires thousands of calls.

See the limits

The limits appear in the Rate Limits tab of System › API (/dashboard/api?tab=rate-limits).

Next to the status (Enabled or Disabled), the screen says whether the organization uses the Platform default configuration or a Custom configuration. The limits come in three groups:

Group	What counts	Platform default
GET Requests	Reads	200 per minute and 5,000 per hour
Write Requests	POST, PUT, PATCH and DELETE	60 per minute and 1,000 per hour
Burst	Short bursts of requests	20 requests in 5 seconds

Need different limits? Contact Kobana support: each organization's limits are adjusted by the Kobana team, not through the dashboard.

When the limit is exceeded

The API responds with 429 Too Many Requests, with the Retry-After header saying how many seconds to wait. The body carries the same value in retry_after:

{
  "error": { "code": "TOO_MANY_REQUESTS", "message": "Rate limit exceeded" },
  "retry_after": 42
}

To handle this well in your integration:

1. When you receive 429, wait the seconds in Retry-After before trying again.
2. Don't repeat the call in an immediate loop: refused attempts also count and only extend the block.
3. Spread large jobs over time — for example, use the API's pagination instead of many parallel fetches.

Each token has its own count: one integration's excess doesn't use up the others' limit. More details in the API reference.