Rate limits

2mee applies limits per connection so one noisy source can’t starve others or the engine.

What’s limited

• Request rate — requests per second per connection/key.
• Payload size — request bodies are capped at 256 KB; larger bodies are rejected with 413 before parsing.
• Burst quotas — short bursts above the steady rate are absorbed up to a ceiling, then throttled.

Exact numbers depend on your plan; your 2mee contact will confirm them. Treat the limits as generous for normal traffic and design for backoff rather than for a specific ceiling.

Handling 429

When you exceed a limit you get 429 Too Many Requests with a Retry-After header:

HTTP/1.1 429 Too Many Requests
Retry-After: 2

• Wait at least Retry-After seconds before retrying.
• Back off exponentially with jitter if you keep getting 429.
• Use a stable messageId so retries are idempotent — a retry after a 429 can never double-process.

Smoothing traffic

• Prefer steady delivery over large bursts where you control timing.
• Batch where a provider supports it (e.g. Segment’s batch endpoint) rather than many tiny requests.

On the outbound side

When 2mee calls your endpoints, it honours your Retry-After and backs off the same way — so a 429 from you is respected, not hammered. See retries & idempotency.