Not Everything Speaks HTTP
Your database, your game server, your custom TCP protocol, your internal RPC service none of it responds to an HTTP health check, but all of it can go down just as easily as a web server can. If the only thing you’re monitoring is your public-facing site, everything behind it is a blind spot.
Seabhac.io probes any TCP or UDP port directly, on any host, so reachability and connection latency are tracked for services that were never meant to speak HTTP.
How it works:
- Point at any host and port: No assumptions about protocol just a socket connection attempt.
- Measure the connection: Track whether the connection succeeds and how long it took, on every run.
- Alert on failure or slowdown: Get notified the moment a port stops responding, or connection times start creeping up.
Where This Actually Gets Used
- Databases Postgres, MySQL, Redis, and anything else listening on a raw port, checked independently of the application in front of it.
- Game servers verify player-facing ports stay reachable, including UDP-based protocols.
- Internal and custom services message queues, RPC endpoints, IoT gateways, anything with a socket.
Reliable Enough to Trust
Connection attempts use progressive retries before firing an alert, so a single dropped packet on a UDP probe doesn’t page your team at 3am for nothing.