Work out uptime and reboot count instantly with clear inputs, formula shown and shareable results.
Long uptime is often presented as a virtue but usually means unactivated kernel patches: an installed kernel does not take effect until reboot, so a 400-day uptime on a patched system means hundreds of days running vulnerable code. Reboots cost a few minutes each, which is negligible against the exposure they close.
Reboot impact
downtime = reboots x minutes per reboot; availability = 1 - downtime / period; mean uptime = period / reboots
Not for security. It usually indicates deferred kernel updates. Regular planned reboots, ideally rolling across a cluster, are healthier than long uptime.
Live patching exists for critical fixes on some distributions, but it covers a subset of changes. Reboots remain necessary for full kernel upgrades.