PrincipalKubernetes

When would you run multiple Kubernetes clusters instead of one big one, and how do you manage a fleet?

What they are really testing: Principal-level: do you weigh isolation and blast radius against operational complexity, and resist multi-cluster unless there is a real driver. They want judgment plus how you would operate it.

A real interview question

When would you run multiple Kubernetes clusters instead of one big one, and how do you manage a fleet?

What most people say

drag me

I would run separate clusters for each environment and team so they are isolated.

A cluster per team/env can be the right call, but stated as a reflex it ignores the heavy operational cost and that namespaces handle much isolation. A principal weighs the trade-off and has a fleet-management answer.

The follow-ups they ask next

  • What is the strongest argument against multi-cluster?

    Operational cost and complexity: upgrades, monitoring, policy, and networking all multiply per cluster, and clusters drift into snowflakes without strong automation. Much isolation is achievable with namespaces/RBAC/quotas in one cluster.

  • How do you keep many clusters consistent and avoid drift?

    GitOps (Argo CD/Flux): cluster state is declared in git and continuously reconciled, so config is consistent and auditable across the fleet, plus uniform policy via admission control (Kyverno/Gatekeeper).

What the interviewer is listening for

  • Defaults to fewer clusters + namespace isolation
  • Names real drivers (blast radius, region/DR, compliance, scale)
  • Manages the fleet with GitOps + uniform policy

What sinks the answer

  • Reflexively a cluster per team/env
  • Ignores the operational cost
  • No fleet-management/anti-drift answer

If you genuinely do not know

Say this instead of freezing. Reasoning out loud from what you do know beats silence every single time, and a good interviewer is listening for exactly that.

Default to [fewer, bigger clusters, since namespaces/RBAC/quotas/NetworkPolicies give strong isolation]. Go multi-cluster only for [real drivers: blast-radius/upgrade isolation, region/DR, hard compliance or tenant isolation, scale limits]. Then run it [as a GitOps-managed fleet (Argo CD/Flux) to prevent drift], and solve [cross-cluster networking/mesh, central observability, uniform policy].

Keep going with kubernetes

All 336 cloud engineer questions

Knowing the answer is not the same as recalling it under pressure

Sign in to send the questions you fumble to spaced recall, so they come back right before you would forget them, and learn the concepts behind them with hands-on labs.

Start free