Practise with 617 DevOps interview questions across Linux, Kubernetes, Terraform, CI/CD and production operations. Start with a worked answer, choose a preparation path, then use the topic previews to find your next exercise. Each full answer includes an approach, an explanation, follow-up questions and common mistakes.
Grounded in researched DevOps, SRE and platform engineering interview loops, written to a senior-engineer editorial bar, and never padded to hit a word count.
These openings come from the existing worked answers. Each link opens the full explanation and follow-up, with no account required.
TL;DR: A VM runs a full guest kernel on virtualised hardware. A container is an ordinary Linux process whose view is narrowed by namespaces and whose resources are capped by cgroups, sharing the host kernel. Sharing avoids booting a guest kernel, but startup still includes image pulls and application initialization. The shared kernel also shapes the security boundary.
TL;DR:
initdownloads providers and connects to the backend,plancompares your code against state against reality and prints the difference,applyexecutes that difference,destroyremoves what state says it owns. Always save the plan to a file and apply that file, or you approved one thing and ran another.
TL;DR: Commit, build once into an immutable artifact, test in widening circles, scan, publish, then deploy that same artifact through environments with a gate at each. The part that separates answers is naming what fails each stage and what the whole thing costs in minutes.
TL;DR: TCP is a connection with ordering, retransmission and flow control; UDP is a datagram with none of that. You choose UDP when a late packet is worth less than a lost one, or when you want to build your own recovery, which is exactly what QUIC does.
Use your target responsibilities and current gaps to choose a path. These are our study recommendations; a job title or years of experience alone does not determine an interview format.
Trace a request, inspect a Linux service and explain how a commit becomes a deployment. Use a small lab and record the commands and observations.
Start with service diagnosis →Explain a failed rollout, an infrastructure change and a recovery decision. Name the evidence, permissions and rollback conditions for each.
Work through a drift decision →Set a user-facing objective, calculate capacity and remove a failure domain. State the assumptions your design needs and how you would test them.
Calculate recovery capacity →Use five sessions, then revisit the weakest result. Each session should produce something you can explain or inspect.
Select exercises that match the vacancy. There is no universal weighting for these areas; use the expected work to decide where to go deeper.
Distinguish an observation from a hypothesis. Explain the next check, the risk of a proposed action and what would establish recovery.
State capacity, availability and recovery requirements. Calculate a bound with units, then test a failure scenario.
Trace what the system does: Terraform state reconciliation, Kubernetes scheduling or a Linux process starting. Use commands and evidence to support the explanation.
Test a useful script with empty input, malformed data, partial failure and repeated execution. Ask separately whether the assessment includes algorithms.
Explain a decision you made, its tradeoff and its result. Label unknowns and distinguish your work from the team’s contribution.
Signals and the OOM killer, cgroups v2 and namespaces, systemd, /proc, TCP states and TIME_WAIT, DNS resolution order, and the Bash/Python/Go a Google SRE or Meta PE screen actually asks you to write.
IAM least privilege, VPC design, security groups versus NACLs, why peering is not transitive, Transit Gateway routing, multi-account landing zones and multi-region failover across AWS, Azure and GCP.
Terraform and OpenTofu state, remote backends and locking, drift detection, count versus for_each, module design, lifecycle blocks, Ansible idempotence, and the BUSL license split every interviewer now probes.
Control plane internals, the full kubectl apply to running pod path, scheduling and affinity, native sidecars, in-place pod resize, CNI and network policy, CSI and StatefulSet identity, RBAC, operators and service mesh.
Pipeline design and caching, artifact promotion, rolling versus blue-green versus canary and what each costs, progressive delivery with Argo Rollouts and Flagger, ArgoCD and Flux reconciliation, and the DORA metrics behind them.
Metrics versus logs versus traces, PromQL, Prometheus cardinality blowups, OpenTelemetry collectors and temporality, exemplars, SLI and SLO design, error budget math, and burn-rate alerting that survives an on-call rotation.
The highest-signal round on the loop. CrashLoopBackOff with empty logs, exit code 137, ImagePullBackOff, stuck PVC finalizers, 503s from a healthy-looking service, p99 tails, incident command, and the blameless postmortem.
SAST, DAST, SCA and IAST placement in a pipeline, Trivy and Checkov gates, secrets management and rotation with Vault, image signing and SLSA provenance, zero trust in Kubernetes, and compliance as code.
Google-style NALSD with napkin math before boxes, multi-region active-active, disaster recovery under a real RTO, capacity planning, blast radius, and designing a delivery platform for hundreds of services.
Internal developer platforms, golden paths, Backstage catalogs and templates, Crossplane, self-service without a ticket queue, plus cost allocation, showback, Karpenter, spot strategy and commitment management.
The fastest-growing DevOps sub-domain: GPU scheduling with MIG, time-slicing and DRA, vLLM paged attention and continuous batching, KServe and Ray, gang scheduling with Kueue, and the cost model behind inference.
Amazon Leadership Principles in STAR form, the outage story that shows ownership rather than blame, pushing back on a bad deadline, driving a migration nobody wanted, and explaining risk to people who do not write code.
Company guides combine preparation topics with the available hiring evidence. The reviewed Google, Razorpay, Accenture, Infosys and TCS guides include scoped sources and original exercises. Confirm your own role’s schedule with the recruiter; other guides still carry qualified historical notes.
Start with the responsibilities in the vacancy. This page offers Linux, networking, cloud, infrastructure-as-code, delivery, troubleshooting and design practice. Ask your recruiter which skills are assessed and choose the matching path below.
Use the five-session plan on this page: map the role, diagnose a failure, automate a change, size a system and rehearse an explanation. Spend extra time on the gaps revealed by those exercises. A week is a focused review window, not a promise of complete preparation.
Practise state recovery, deployment compatibility, access boundaries and incident decisions. Explain the evidence behind a change, its failure modes and how you verify recovery. Connect the exercise to a project you owned without inventing production experience.
Responsibilities overlap, but titles do not establish a universal interview format. Compare the vacancy’s software, operations and reliability responsibilities. Use the SRE and design paths when those responsibilities apply, and confirm the actual exercises with the recruiter.
Prepare coding or scripting when the role includes automation or software development. Practise parsing data, handling failures and making a change safe to rerun. Ask about the language, environment and whether algorithmic exercises are included.
The first 18 answers in each topic are readable without an account. A free account opens the first 28. Later answers require Premium. The worked samples on this page come from the anonymous free allowance.
The online question bank links to individual worked answers. Four Premium PDF guides cover incident response, Kubernetes rollouts, CI/CD automation and observability. Their guide pages include free sample pages; they are separate from this question list.
617 questions across 12 topics, ordered easy to hard, with worked explanations, follow-ups and mistakes to avoid. The first 18 answers in every topic are readable without an account.
Loops get revised constantly and differ by team, by level and by region. Prepare in this direction, then ask your recruiter what the rounds actually are.
