Cloud & DevOps: interview questions and learning guide

Linux and shell basics

The terminal is a very fast steering wheel. You must learn what each turn does.

Most servers run Linux, and engineers work through the command line. Core skills are moving through folders, reading files, searching with grep, checking processes and disk, and editing configuration. Permissions decide who can read, write or run files. Services are started and checked with the system service manager, and logs show what went wrong. Scripts and scheduled jobs automate repeated tasks. Careful habits such as checking before deleting prevent outages.

Interview tip: Show that you read logs first and change one thing at a time.

CI/CD and DevOps practices

DevOps is making the road from code to customer short, safe and boring.

DevOps joins development and operations so software is built, tested and released in small safe steps. Continuous integration builds and tests every change. Continuous delivery prepares the release and deployment can be automatic. Small releases, version control, automated checks and quick rollback reduce risk. Monitoring and alerts tell the team when something breaks, and blameless reviews turn failures into fixes.

Interview tip: Say how you would roll back, not only how you would deploy.

Containers and infrastructure as code

If the setup lives only in someone's memory, it is a risk. Put it in code.

Containers package an application with what it needs so it runs the same everywhere. Images are built from a recipe and run as containers. An orchestrator such as Kubernetes starts, restarts and scales many containers. Infrastructure as code describes servers, networks and storage in files, so environments can be created and reviewed like software. Cloud providers rent computing by use, so cost control and security settings matter from day one.

Interview tip: Mention cost and security together when you speak about cloud design.

A commit triggers a build and automated tests, then the artifact is deployed to a test stage, and after checks to production, with monitoring and a way to roll back.

An image is the saved, read-only recipe. A container is a running instance started from that image.

Check the status and logs, find recent changes, restore service first, often by rolling back, communicate to stakeholders and review the cause afterwards.

It makes environments repeatable and reviewable, keeps changes in version control and removes manual drift between environments.

IaaS gives virtual machines and networks, PaaS gives a platform to run code, and SaaS gives a finished application.

A container shares the host operating system kernel and is light and fast. A VM carries a full guest operating system.

Continuous integration builds and tests every change automatically. Continuous delivery deploys it to environments with little manual work.

Give every user and service only the permissions needed for the job, and no more.

It spreads incoming traffic across several servers so no single one is overloaded and the service stays available.

Use top for CPU and memory, df for disk, iostat for disk activity and logs for errors, then find the process behind the load.

Describing servers and networks in files such as Terraform so they can be versioned, reviewed and recreated reliably.

Vertical scaling makes one machine bigger. Horizontal scaling adds more machines behind a load balancer.