Container Security and Its Limits
Containers package and isolate processes but share a host kernel; privileged mode, host mounts and broad capabilities erase much of the boundary.
Frame the problem
Security starts with a concrete asset, attacker capability and trust crossing.
Why the system fails
The workload runs privileged, mounts host control sockets, uses a writable image, or carries permanent secrets.
The important question is not “what is Container Security and Its Limits?” but “which assumption let untrusted data or an over-scoped identity cross container namespace/cgroup → shared host kernel?” Trace the decision at the boundary, then constrain what can happen after the first control fails.
Design the control in layers
Start with the control closest to the interpretation or privilege boundary: Run rootless/non-root and drop capabilities Then add a control that reduces blast radius and telemetry that proves the decision was enforced.
The resulting design is not labelled secure. Record the identified controls, the known failure paths, the remaining exposure, and the evidence you would need during an incident.
| Prevent | Detect | Recover |
|---|---|---|
| Run rootless/non-root and drop capabilities · Avoid host mounts and privileged mode · Use read-only filesystems, resource limits and signed minimal images | Runtime behavior outside the image baseline and host-sensitive access | Contain the affected identity or component, scope impact from audit evidence, and preserve a regression test. |
Key points
- Asset: Host kernel, container neighbors, image provenance and workload secrets.
- Boundary: Container namespace/cgroup → shared host kernel
- Primary control: Run rootless/non-root and drop capabilities
- Detection signal: Runtime behavior outside the image baseline and host-sensitive access
- Always ask what limits damage when the primary control fails.
Boundary control exercise
This lesson uses the shared boundary-control exercise.
Follow the attack
Safe conceptual simulation: capability → missing control → crossed boundary → asset impact.
- 1Attacker starts with: A compromised container process or malicious image.
- 2The workload runs privileged, mounts host control sockets, uses a writable image, or carries permanent secrets.
- 3The weak or missing boundary control is crossed: Container namespace/cgroup → shared host kernel
- 4Impact: Container breakout, host control or lateral compromise.
- Container breakout, host control or lateral compromise.
Defend, detect, recover
One prevention is a single point of security failure. Layer it and make failure observable.
- • Run rootless/non-root and drop capabilities
- • Avoid host mounts and privileged mode
- • Use read-only filesystems, resource limits and signed minimal images
- • Runtime behavior outside the image baseline and host-sensitive access
- • Contain the affected identity or component.
- • Scope access from audit evidence.
- • Fix the boundary and add a regression test.
- • Misconfiguration and new access paths can bypass the intended control.
- • A privileged insider or compromised control plane may still reach the asset.