Guide

Rate limits

Limits protect actions that send mail, contact your systems or move a lot of data. When you hit one, the API answers 429 with a Retry-After header and retry_after_s in the body.

429 · application/problem+json

HTTP/1.1 429 Too Many Requests
Retry-After: 1260

{
  "type": "https://cloudraw.com/problems/rate-limited",
  "code": "rate-limited",
  "title": "rate limited",
  "status": 429,
  "detail": "At most 10 exports an hour per admin; retry in 1260 s.",
  "retry_after_s": 1260
}

Current limits

What	Limit	`code`
Audit log export (`GET …/audit/export`)	10 per hour per token, per workspace	`rate-limited`
Connector commands (update now, turn on updates, pause, resume)	10 per 10 minutes per workspace (all callers together)	`too-many-connector-commands`
Bulk invite (`POST …/people:bulk-invite`)	10 per hour per workspace, up to 200 people each	`too-many-bulk-invites`
Test deliveries (webhook, log stream and alert-channel tests)	10 per hour per workspace and kind	`test-rate-limited`

Each operation with a limit shows a rate limited badge and the limit in its description.

Good citizenship

Keep to a few requests per second per token, and run bulk jobs one request at a time. Lists such as people, devices and connectors return the whole workspace in one call, so you rarely need many requests.
Prefer webhooks or log streaming over polling. If you poll, every 30–60 seconds is plenty.
Always handle 429 and 503 by waiting Retry-After seconds. Cloudraw may add a general per-token limit in the future; clients that honour Retry-After will keep working unchanged.