Vultr Load Balancers perform health checks on backend instances to ensure traffic is routed only to healthy servers. If health checks are failing, consider the following:
- Ensure all backend instances are running and reachable. Instances that are powered off, misconfigured, or unresponsive will appear unhealthy.
- Confirm that the Load Balancer is checking the correct port and that backend services are listening on that port.
- Verify that the health check protocol (HTTP, HTTPS, or TCP) matches the application’s behavior. TCP checks require a listening service, while HTTP/HTTPS checks must return a valid status code.
- For HTTP or HTTPS checks, ensure the configured path responds with a
200–299 status code. Redirects, errors, or empty responses are treated as unhealthy.
- Check that Firewall Rules or network policies do not block the Load Balancer from reaching backend instances on the health check port.
- Evaluate Health Check intervals, timeouts, and thresholds. Applications with slow startup times or temporary high load may be marked unhealthy if thresholds are too strict.
Following these steps helps isolate the cause of health check failures and ensures that traffic is consistently routed to healthy backends.