Server API
Errors and limits
How the Server API answers when something goes wrong, and how much you can call it. The same rules apply to v1 and v2.
#Errors
Every error answers with JSON like {"error": {"code": "...", "message": "..."}}. The message says what went wrong in plain words.
| Status | Code | What it means |
|---|---|---|
| 400 | invalid | Something in the request is wrong. The message says what. |
| 401 | missing_key, bad_key | The key is missing, wrong or revoked. |
| 403 | not_admin, revoked | The key's developer is no longer an admin of the server or lost Server API access. |
| 403 | wrong_version | A v1 key was used on v2. |
| 404 | not_found | No such thing, or no such endpoint. |
| 409 | The change can't happen right now, like starting a second session. | |
| 429 | rate_limited | Wait for the seconds in the retry-after header. |
#Limits
- 60 requests a minute per key. With a v2 key, 20 of those can be changes.
- Every answer has
x-ratelimit-limit,x-ratelimit-remainingandx-ratelimit-resetheaders. - Up to 10 keys at once, and 3 webhooks per key.
- Keep your key on a server or bot. Calls from web pages are refused.
- Every call is logged for 30 days. Uplink owners and the server's admins can see and revoke your keys.