Live Servers
Open Analytics → Telemetry to inspect servers running your application. Connect the agent using the setup instructions available from Telemetry; the same integration supplies load-test server metrics.
Each server can report CPU, process memory, sessions, memory per session, threads, and managed heap metrics where its framework supports them. Multiple servers can report to the same project.
Live metrics stream while a load run is active or the view is watched. The agent still polls for work while idle, and application errors and lifecycle events can be reported independently. Telemetry is not all transient: sampled history and errors are retained and follow Settings → Storage → Telemetry retention.
A server with no recent samples is not proof of an idle or healthy application. Check its last-seen state, agent configuration, and network access.