Server monitoring
The server list shows latency, health and traffic. Refreshing is on demand: you get current data when you open the view, while tray monitoring drops to a lower frequency so the monitoring tool does not itself become the background load.
Bind VPS traffic with one command
Server monitoring can bind a Linux VPS to your account and track traffic per billing cycle. Binding is light: run a single install command on the target machine, and inbound and outbound traffic accumulates from there without any manual bookkeeping.
This is what stops the classic "throttled at the end of the month before you notice you went over" — usage stays visible instead of arriving with the bill.
Cycles, reset dates and provider calibration
Billing cycles differ per provider, so reset rules are configurable: choose monthly or yearly resets and set the reset month and day. More usefully, calibration lets you align the current cycle to the number shown in your provider's panel when the two drift apart, so later figures track the real bill instead of restarting from zero.
What the status colours mean
Latency colours were redefined to avoid alert fatigue from a wall of red: high latency is orange, and dark red is reserved for genuinely unreachable servers. VPS cards use a denser layout with details and actions moved into hover hints and the context menu, keeping long lists readable. Country flags make regions easy to scan.
Hardware monitoring (Windows desktop)
Hardware monitoring reads local CPU and GPU temperature and power. The tray shows auto-detected short model names with live readings, and a dashboard widget links through to the full monitoring page.
Low-level sensors and signed drivers
Reading low-level sensors requires driver support. 77Fox verifies the official driver's version, hash and digital signature before enabling it, then restarts the sampling process automatically. Production installs handle this on first launch or after an upgrade, and Windows builds run elevated by default to reach low-level sensors.
If low-level sensors cannot be enabled — for example when a source-build process is not running elevated — the app states the status clearly and offers a re-detect and repair action. GPU monitoring is unaffected. The web console does not run local hardware monitoring.
History and electricity cost
The full page charts temperature, power and energy over ranges from one hour to 30 days. Set your electricity price in a major currency and the app estimates cost for today and for any range you select. For a machine that runs long jobs, that is a direct answer to "what does this setup actually cost to run per month".