TCP Port Monitors

What a TCP Port Monitor Is

A TCP Port Monitor is an active reachability check that works below the HTTP layer. LoadFocus opens a raw TCP connection to a host and port from each selected region, measures the connection time, and reports the result as up or down. When the connection succeeds but is slow, the check can also report degraded if the connect time exceeds a threshold you set.

This covers services that an HTTP check can't reach — anything that listens on a TCP port but doesn't speak HTTP.

When to Use It

• Databases — PostgreSQL (5432), MySQL (3306), Redis (6379).
• Mail servers — SMTP (25, 587).
• Custom TCP services — game servers, message brokers, internal daemons.
• Raw port reachability — confirm a port is open and accepting connections, independent of any application protocol.

Create a TCP Monitor

1. Open the New API Check page.
2. Choose the TCP type.
3. Enter the host (hostname or IP address).
4. Enter the port — any value from 1 to 65535.
5. Set the connection timeout in milliseconds — how long to wait for the TCP handshake before marking the check down.
6. Optionally set a degraded threshold in milliseconds — if the connection time exceeds this but still succeeds, the check is reported as degraded rather than fully up.
7. Pick your locations, alert channels, and frequency, then save.

Multi-Region Checks

Like other LoadFocus checks, a TCP monitor can run from multiple regions. Each region opens its own connection and reports its own connection time, so you can see whether a port is reachable everywhere or only from certain parts of the world. If the connection fails in any selected region, the check is marked failed.

For the shared settings — frequency, locations, alert channels, activate/mute — see How to Create a New API Check.

You can find all your checks on the API Monitors page.