Recipe

Monitor public-API health from your side

Watch the public API from your side with a light touch. A periodic GET on a cacheable endpoint like the loyalty tiers is a fine liveness probe — just keep the interval well under the rate limit and treat a 200 with a fresh generated timestamp as healthy. Never health-check by hammering a write endpoint.

2 min read

GETProbe reads, not writes
≪60/60 sStay under the limit
200 + freshHealthy signal

Pick a read probe

Use a cheap, cacheable read — the loyalty tiers endpoint is ideal. A single request every minute or two is plenty and stays far under the 60/60 s window.

Define healthy

Healthy is a 200 with a well-formed body and a recent generated timestamp. A 429 means you are probing too hard, not that the API is down — back off. A 5xx or a timeout is a real signal.

Do not probe writes

Never health-check by posting enquiries or consent — that pollutes real data and burns your rate budget. Probe reads only.

Frequently asked questions

Which endpoint should I use for uptime checks?

A cacheable read like GET /public/v1/loyalty/tiers. It is cheap, carries no PII, and a 200 with a fresh timestamp is a clean healthy signal.

My probe returns 429 — is the API down?

No. A 429 means your probe is calling too often for its IP. Increase the interval; the API is fine.

Funding for UK limited companies

Credicorp lends to your company, not to you personally — short-term working capital with no personal guarantee. See what your business could access.