It’s 180 DevOps interview questions with short, say-it-out-loud answers, grouped by the tool the interviewer is holding. Culture and lifecycle, Linux, Git, CI/CD and Jenkins, Docker, Kubernetes, Terraform, AWS, and the monitoring and troubleshooting rounds. One question per block, one tight answer under it, with real commands where a command is the answer. Built for anyone with a DevOps or platform engineer screen booked in the next two weeks.
Yes, this DevOps interview questions PDF is a free download. Drop your email below and it lands in your inbox. No payment.
What this PDF is
It’s 180 DevOps interview questions with short, say-it-out-loud answers, grouped by the tool the interviewer is holding. Culture and lifecycle, Linux, Git, CI/CD and Jenkins, Docker, Kubernetes, Terraform, AWS, and the monitoring and troubleshooting rounds. One question per block, one tight answer under it, with real commands where a command is the answer.
Most DevOps prep is a wall of text. You open it, read 40 pages, and still freeze when someone asks you to explain CMD versus ENTRYPOINT out loud. Knowing a thing and being able to say it in fifteen seconds are two different skills, and only one of them gets you the offer.
This trains the second one. Every answer is one or two sentences, phrased the way you’d say it across a desk. Read it on the train. Cover the answers, quiz yourself, move on. It’s a cheat sheet, not a textbook.
There’s a reason it’s worth the effort. DevOps ranked fifth in job demand for 2026, and 81% of organisations have adopted DevOps practices or plan to (HackerX, Programs.com). Median pay sits around $182K on 2026 job-posting data (Recruiting from Scratch). The competition for those seats is real, and the screen is where most people lose. Thinking past the exam? The 200-901 DevNet skills that get you hired shows what employers ask for.
What’s inside
- DevOps fundamentals – what DevOps actually is, why it exists, the five core principles, the eight lifecycle stages, CI versus CD versus continuous deployment, DevOps versus Agile, and the one-line answers that stop an interviewer digging
- Deployment and reliability – blue-green, canary, rollback, immutable infrastructure, shift-left testing, DevSecOps, SRE versus DevOps, plus SLA, SLO, SLI and MTTR explained so you don’t mix them up under pressure
- Linux for DevOps – kernel and distros, the directory structure they make you recite, root versus sudo versus su, checking CPU, memory and disk, processes and zombie states, systemd and daemons, cron syntax, logs, swap, inodes, soft versus hard links, SELinux modes
- Git and version control – the three areas, add versus commit, fetch versus pull, merge versus rebase and when never to rebase, resolving conflicts, stash, reset versus revert, cherry-pick, HEAD, centralised versus distributed
- CI/CD and Jenkins – architecture, controller versus agent, freestyle versus pipeline, declarative versus scripted, a working Jenkinsfile, stages, plugins, webhooks, credentials and secrets handling, artifacts, retries, parameterised builds, shared libraries, and how to debug a red build
- Docker – images versus containers, VM versus container, namespaces, cgroups and UnionFS, every Dockerfile instruction, CMD versus ENTRYPOINT, volumes and the three storage types, networking modes, EXPOSE versus publish, Compose, multi-stage builds, image size and security practices
- Kubernetes – control plane and worker components, API server, etcd, scheduler, controller manager, kubelet, pods, deployments, ReplicaSets, all four service types, Ingress, ConfigMaps and Secrets, liveness versus readiness probes, rolling updates, HPA, namespaces, DaemonSets, StatefulSets
- Terraform and IaC – the init-plan-apply-destroy workflow, providers, state and why you never delete it, local versus remote backend and state locking, variables, outputs, modules, workspaces, lifecycle rules, taint, import, Terraform versus Ansible
- AWS and cloud – IaaS versus PaaS versus SaaS, EC2, AMIs, security groups, VPC, public versus private subnets, IAM and why roles beat users, S3, load balancer types, auto scaling, CloudWatch
- Troubleshooting rounds – the five-step Kubernetes debug sequence, the Jenkins failure checklist, and the common traps that get candidates rejected Latency, traffic, errors, saturation, the 4 golden signals from Google’s SRE book. If an interviewer asks what you’d monitor with only one dashboard, this is the answer.
Every answer is short on purpose. If they want depth, they’ll ask a follow-up. A two-sentence answer beats a two-minute ramble every time.
Who it’s for
- Career switchers coming from sysadmin, support or dev who need the vocabulary fast
- Junior and mid-level DevOps applicants with a first-round technical screen booked
- Cloud and platform engineers brushing up on the tools they touch least
- Students and self-taught engineers who’ve built the labs but never said any of it out loud
It sits at junior to mid level. Fundamentals through Kubernetes and Terraform, with the senior system-design rounds left out on purpose.
What should I study after this?
The hands-on half. Reading answers gets you through the screen. Building a pipeline gets you through the offer. Our DevOps Lab Workbook covers that hands-on.
What the Interview Loop Actually Looks Like
Most DevOps loops run through 5 rounds, each testing something different.
| Round | What They’re Testing | Typical Time |
|---|---|---|
| Recruiter screen | Story, salary, notice period | 20 min |
| Technical phone | Fundamentals, tool depth | 45-60 min |
| Live troubleshooting | Real broken pipeline or cluster | 60-90 min |
| System design | Build a CI/CD or infra platform | 60 min |
| Behavioural / on-call culture | Incidents, postmortems, disagreements | 45 min |
5 Mistakes That Cost Candidates the Job
- Memorising definitions instead of understanding tradeoffs.
- Not having a story.
- Freezing on troubleshooting rounds.
- Dismissing soft questions.
- Talking about tools, not outcomes.
For behavioural questions, use STAR: Situation, Task, Action, Result. Keep it short. The win is showing you can disagree without drama and end up in a better place than where you started.
For the “tell me about yourself” opener, three parts, 90 seconds total: who you are now (current role, stack), one signature story (a real thing you shipped with a measurable result), and why you’re excited about this specific company.
