In Kubernetes and similar cloud-native systems, a liveness probe is an automated health check that answers a narrow question: is this running service still healthy enough to keep going? If the check fails, the platform restarts the container or process. This is different from a readiness check, which decides whether a service should receive traffic.
Liveness probes matter in cyber security because they help recover from hangs, deadlocks, memory exhaustion, and other failure states that attackers can trigger with malformed requests or resource abuse. They are also a defensive control against quiet service degradation: a system may still be online while its logic is stuck or broken. But probes must be designed carefully. If an attacker can cause repeated probe failures, they may create restart loops and denial of service. Good operators pair probes with monitoring, logs, and limits so restarts improve resilience instead of masking compromise.