If an exec liveness probe calls a command that is missing from the container image, Kubernetes cannot run the check. The probe fails; after enough consecutive failures, the kubelet treats the container as unhealthy and restarts it. The Kubernetes documentation describes this behavior, though it does not verify a particular production incident matching this headline.
Why a missing command can trigger restarts
An exec probe runs its configured command inside the container. Kubernetes considers the check successful only if the command exits with status 0. The executable therefore needs to exist in the final image, be usable, and be available at the configured path or through the execution environment’s PATH. See Kubernetes: Liveness, Readiness, and Startup Probes.
As an Amazon Associate I earn from qualifying purchases.
Do not assume Kubernetes invokes a shell. The command is executed as configured; if shell behavior is required, the probe must explicitly run a shell and pass it the command.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →When the executable cannot be launched, the exec check fails. A Kubernetes issue report includes the runtime message executable file not found in $PATH, but that is an example diagnostic, not a guaranteed message for every runtime or failure. Kubernetes issue #102594
#1 Best Overall
Kubernetes summarizes the purpose directly: “Liveness probes determine when to restart a container.” A missing binary is not itself proof that the application process is broken; it means this configured check could not succeed.
What happens after the probe fails
For liveness and startup probes, reaching the configured failureThreshold causes Kubernetes to treat the container as unhealthy and restart it. Readiness has a different job: failed readiness marks the container unready so it is excluded from Service traffic, without restarting it solely because of that failure. Kubernetes probe documentation
| Probe type | What failure means | Typical purpose |
|---|---|---|
| Liveness | At the failure threshold, the container is treated as unhealthy and restarted. | Detect a condition a restart can plausibly fix. |
| Readiness | The container remains running but is marked unready and removed from Service traffic. | Determine whether the container should receive traffic. |
| Startup | At the failure threshold, the container is treated as unhealthy and restarted; while it is active, it gates liveness and readiness checks. | Allow initialization to finish before other probes begin. |
The documented defaults are a failureThreshold of 3 consecutive failures, a periodSeconds of 10 seconds, and a timeoutSeconds of 1 second; the failure threshold and timeout have a documented minimum of 1. These are Kubernetes configuration defaults, not guaranteed values for a particular workload: check the probe configuration in your manifest. Kubernetes: Configure Liveness, Readiness and Startup Probes
Increasing thresholds only changes when Kubernetes acts. It does not make an absent executable runnable, so threshold tuning is not a fix for a missing binary.
How to diagnose the restart loop
- Read the probe configuration. Inspect the affected container’s
livenessProbe.exec.commandexactly as written. Look for the executable, arguments, and any explicitly invoked shell. - Check the final image. Verify that the executable is included in the image actually deployed, has suitable permissions, and is reachable at the configured path or through the relevant
PATH. A utility available in a build stage or a developer’s local environment may not be present in the runtime image. - Inspect Pod events and container state. Look for
Unhealthyevents, liveness failure messages, runtime errors, and changes to the container restart count. The Kubernetes probe tutorial demonstrates checking Pod events after probe failures. - Check whether liveness is testing the right condition. A liveness check should identify a problem the process can recover from by restarting. A temporary dependency outage or high load may make an application unhealthy without making a restart helpful.
- Choose a probe that fits the need. Use readiness when the question is whether the container should receive traffic. Use a startup probe when initialization needs time before liveness or readiness checks begin. Consider HTTP, TCP, or gRPC probes if they can test the intended condition without relying on a utility binary in the image.
Choose checks that help rather than amplify failures
Kubernetes warns that “Incorrect implementation of liveness probes can lead to cascading failures.” If a probe treats transient load or a shared downstream outage as a reason to restart containers, repeated restarts can compound an existing problem rather than resolve it. Keep liveness focused on failures for which restarting is an appropriate recovery action. Kubernetes: Configure Liveness, Readiness and Startup Probes
Exec probes also launch processes. Kubernetes notes that frequent exec probes in dense clusters may add CPU overhead. When selecting a mechanism, weigh whether it can verify the intended health condition, whether it depends on a binary in the image, whether its failure should affect traffic or trigger a restart, and the overhead at your probe interval and Pod density. Kubernetes probe documentation
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
