TCP port and network service monitoring
Monitor whether a database, gateway or custom network service is reachable on the expected port. RMON checks connectivity from selected locations and alerts when the service stops responding.
Check the service endpoint your applications depend on
A host can answer Ping while an important application port is unavailable. RMON resolves the service name, attempts the configured TCP connection and records whether a usable endpoint responded.
A successful connection confirms that the port accepts connections. It does not validate the application response. Multi-location checks show whether reachability differs by network path.
Monitor the right port
Check the actual service endpoint instead of relying only on host availability.
Measure connection time
Track connection timing and alert when the response-time threshold is exceeded.
Locate connectivity failures
Compare results and connection timing from different monitoring locations.
What this solution can cover
- DNS-name or IP-based TCP checks
- IPv4 and IPv6 connectivity
- Sequential attempts across resolved addresses
- Name-lookup and connection timing
- Retries and response-time thresholds
- Multi-location alerts and history
Create a TCP service check
- 01
Enter the endpoint
Choose the service DNS name or address and the TCP port it should accept.
- 02
Set the schedule
Choose how often to connect and how long to wait before a timeout.
- 03
Choose locations
Assign agents or regions that use the relevant network paths.
- 04
Set thresholds and alerts
Configure retries, timing limits, notification channels and a runbook.
Common questions about tcp port monitoring
Can RMON monitor any TCP port?
RMON can attempt a connection to the configured TCP endpoint and port, subject to normal network access.
Does a successful TCP check confirm application health?
It confirms that the port accepts a connection. Use an HTTP, SMTP or RabbitMQ check when you need to test the application protocol.
Does RMON support IPv6 TCP checks?
Yes. TCP checks can use IPv4 and IPv6 addresses returned for the service name.