JustaName Status Page

Public uptime monitoring + status page for health.justaname.id, watching api.justaname.id (CCIP-Read gateway, resolution endpoints, bundler RPC) with a 99.9% monthly uptime target.

Built on Gatus. Runs on a DigitalOcean droplet — outside the AWS/EKS infra it monitors, so the status page stays up to report an outage instead of going down with it.

Checks

Group	Check	Endpoint	Asserts
API	liveness	GET /.well-known/x402.json	server up (public, no auth/RPC)
CCIP-Read gateway	resolve	GET /ens/v2/resolve?ens=…	viaUniversalResolver==true, correct ens
CCIP-Read gateway	batch resolve	GET /ens/v2/resolve?ens=a&ens=b	array response, first ens correct
Resolution API	reverse	GET /ens/v2/reverse?address=…&coinType=60	resolves without error
Bundler	base / ethereum	POST /proxy/v1/rpc?chainId=… eth_chainId	returns the right hex chainId

The resolution checks assert correctness, not just a 200: viaUniversalResolver is only true when the request actually went through the Universal Resolver / CCIP-Read path, so a gateway returning 200 with garbage still counts as down.

Cost — effectively $0

• Liveness: static JSON, no auth, no RPC. Free.
• Resolution (resolve/batch/reverse): we pass a free public RPC (PublicNode) via X-Rpc-Url, so the on-chain compute is paid by the public node, not JustaName's dRPC account. No api key either (free BYO-rpc path).
• Bundler: eth_chainId is answered locally by the bundler (never reaches dRPC). The only cost is the api-key validation the guard runs per request — internal load, not money. Runs every 5min to keep that minimal.

So the whole page costs ~$0 in RPC/spend; the only thing to watch is the api key's own quota for the bundler checks.

Intervals

liveness 30s (free, fast full-outage detector) · everything else 5min (the resolution checks lean on a free public RPC, so we don't probe it aggressively).

Deploy

See DEPLOY.md for the full DigitalOcean walkthrough (droplet, DNS, secrets, TLS). Short version:

git clone https://github.com/JustaName-id/status-page.git && cd status-page
cp .env.example .env     # fill in RPC_URL, KNOWN_NAME(_2), KNOWN_ADDRESS, BUNDLER_API_KEY, SLACK_WEBHOOK_URL
docker compose up -d     # Gatus + Caddy (auto-HTTPS for health.justaname.id)

Local preview

cp .env.example .env
docker compose -f docker-compose.local.yml up   # http://localhost:8080, no TLS

Probes still hit prod, so you see real results. History here is throwaway.

Multi-region (for the 99.9% SLA)

Run instances in 2+ regions and confirm "down" from more than one vantage point. For unified history across regions, switch storage to Postgres (commented block in config.yaml).

Files

• config.yaml — Gatus checks, UI, alerting
• docker-compose.yml — Gatus + Caddy (prod, auto-HTTPS)
• docker-compose.local.yml — Gatus only, localhost:8080 (preview)
• Caddyfile — reverse proxy + Let's Encrypt for health.justaname.id
• .env.example — required environment variables