DevOps Interview & Career Prep

DevOps is a set of cultural practices, organizational patterns, and technical capabilities that improve an organization's ability to deliver software reliably and rapidly. Interviewers probe DevOps knowledge to assess whether a candidate understands how modern software teams operate — not just individual tool syntax. Core topics include the three-way handshake between development, operations, and security; measurement via DORA metrics (deployment frequency, lead time for changes, change failure rate, time to restore service); and cultural practices such as blameless postmortems and shared ownership.

An error budget is the allowable amount of unreliability derived from an SLO (Service Level Objective). If your SLO is 99.9% availability, your error budget is 0.1% downtime — approximately 8.7 hours per year. The error budget is consumed by outages, degradations, and planned maintenance. When the budget is depleted, new feature releases are halted until reliability is restored, creating a shared incentive between product and reliability teams. The concept originates from the Google SRE book (Beyer et al.).

The four DORA metrics (from Forsgren, Humble, and Kim's Accelerate research) measure software delivery performance: (1) Deployment Frequency — how often code is deployed to production; (2) Lead Time for Changes — time from code commit to production deployment; (3) Change Failure Rate — percentage of deployments that cause a production incident; (4) Time to Restore Service — how long it takes to recover from a production failure. Elite performers (top quartile) deploy multiple times per day, have lead times under one hour, change failure rates under 5%, and restore service in under one hour.

A virtual machine (VM) includes a full operating system kernel, which is managed by a hypervisor. A container shares the host OS kernel and isolates processes using Linux namespaces (PID, network, mount, UTS, IPC) and cgroups (resource limits). Containers start in milliseconds, have megabyte-level overhead, and are portable across any host running the same container runtime. VMs provide stronger isolation but have gigabyte-level overhead and seconds-to-minutes startup times. In practice, containers and VMs are complementary: containers run inside VMs in most cloud environments.

DevOps interviewers commonly test: process management (how to identify and kill processes, what systemd does, what a zombie process is), file permissions (chmod octal notation, setuid/setgid, umask), filesystem layout (what lives in /var, /tmp, /proc, and why), networking basics (how to check open ports with ss, what /etc/resolv.conf controls, how iptables chains work), and shell fundamentals (exit codes, pipe vs redirection, cron syntax). You don't need to memorize every flag, but you should be able to explain the concepts and describe how you'd diagnose common problems.

Blue/green deployment maintains two identical environments (blue = current live, green = new version). Traffic switches entirely from blue to green at cutover. Rollback is instant — switch traffic back to blue. Canary deployment gradually shifts a small percentage of traffic to the new version (e.g., 1% → 5% → 25% → 100%), monitoring error rates and latency at each step before proceeding. Blue/green is simpler and provides faster rollback; canary reduces blast radius by exposing a small fraction of users to the new version first. Canary is preferred when changes are higher risk or when you need real traffic validation.

Infrastructure as Code means managing and provisioning infrastructure (servers, networks, databases, security groups) through machine-readable configuration files rather than manual processes or interactive UIs. The key benefits are repeatability (the same config produces the same infrastructure every time), version control (infra changes go through code review, can be diffed and reverted), and auditability (the git history is the change log). Terraform (declarative, HCL) and Ansible (procedural for configuration management) are the most common IaC tools. Pulumi uses general-purpose programming languages instead of a DSL. IaC is a prerequisite for reliable CI/CD and GitOps workflows.