Ultimate DevOps and Cloud Interview Guide

Explain using git fork to create a private copy for contributing to a repository with write access disabled, and using git clone for local copies when not contributing.

Learn how git fetch updates your local repository with the latest commits, branches, and tags from a remote; git pull combines fetch with merge or rebase.

Explain when to use git pull versus git fetch, noting that git pull keeps the repo up to date for CI/CD checks, while git fetch with cherry-pick suits experimental work.

Demonstrate git rebase versus git merge with a sample repository, showing how commit order differs when integrating main and a feature branch, and how rebase changes the history.

Explain git merge versus git rebase in interviews: merge preserves full history with a merge commit, while rebase creates a cleaner linear history.

Explain three git challenges faced in a prior role: implementing a modified trunk-based branching strategy, tightening access control, and enforcing git hygiene with dot gitignore, pre-commit hooks, and webhooks.

Respond to a Kubernetes secret accidentally committed to git by revoking it, cleaning history with BFG, rotating credentials, and preventing future pushes with pre-commit hooks and education.

Investigate high disk usage on a Linux instance and locate /var/logs. Prevent 100% utilization by rotating, archiving, or deleting old logs, and cleaning apt caches.

Troubleshoot ssh to an instance by verifying the IP address, PEM file and its permissions, firewall status, security groups on port 22, and the sshd service with systemctl.

Use the Linux find command to locate files over 100 MB for disk cleanup, then preview and delete them (with sudo and -exec) to free space on a VM.

Explore local variables scoped to a shell script, environment variables created with export for global access, and special variables like $?, $0, and $1 for status and arguments.

Explain the OSI model by tracing a client request from browser to server across layers seven to one, including http/https, tls/ssl, cookies, and packet routing.

Learn how dns resolution causes curl to fail with domain names but succeed with ip addresses, and fix it by setting nameserver 8.8.8.8 in resolv.conf.

Discover five day-to-day maven build targets: clean, compile, test, package, and install. Learn how these steps compile code, run tests, package artifacts, and publish to local or remote repositories.

Explain how the Terraform state file acts as the brain, storing resource IDs, metadata, IPs, and last known state to guide plans and updates.

Terraform state locking in remote backends prevents conflicts when two engineers run apply concurrently; the first wins the lock and creates the VPC, the second may fail.

Learn how to store the Terraform state file on premises using remote backends like HashiCorp console or Terraform Enterprise, without cloud accounts, with state locking and drift detection.

Understand the difference between resources and data sources in Terraform, where resources create, modify, or delete cloud provider assets, while data sources read existing cloud information.

Learn the true role of the EXPOSE keyword in a Dockerfile as metadata for port documentation, not for publishing, and how Docker Compose uses it for inter-service communication.

Learn how Docker image caching can hide code changes after a rebuild and how to force a fresh build with docker build --no-cache to reflect updates.

Learn when to forcefully remove a container and how, covering stuck or unresponsive containers, continually restarting containers, and CI/CD pipeline scenarios using docker rm -f and docker ps.

Learn why hardcoding pod IPs fails in Kubernetes as pods are ephemeral. Discover how services use labels and selectors to provide stable communication and avoid outages.

Compare NodePort and LoadBalancer Kubernetes service types, focusing on access scope and external exposure. Learn when to use each based on VPC access and Cloud Controller Manager availability.

Explain deployment strategies used, highlighting blue-green and canary approaches: create a new environment with a load balancer for v2, and gradually roll out traffic while ensuring easy rollback.

Learn how taints on a Kubernetes node marked no schedule can be overridden by pod tolerations, enabling exception scheduling during maintenance or upgrades.

Explain how node affinity, the modern alternative to node label selector, uses required during scheduling and preferred during scheduling to place pods on labeled nodes.

Explain how Kubernetes assigns quality of service by pod resource requests and limits, classifying pods as best effort, Burstable, or guaranteed, guiding eviction during memory pressure.

Diagnose app slowness by enabling debug mode, checking Grafana/Prometheus metrics, and using distributed tracing (Jaeger) to map the request path and identify latency sources.

Explain what AWS NAT is, how a NAT gateway enables private subnets to access the internet, and why it translates source addresses for secure outbound traffic.

Understand what trust policy means in AWS and why it is used. Learn how an IAM role can be temporarily assumed via STS by services like Lambda to access DynamoDB.

Learn cross-account access: enable a lambda in account a to access dynamodb in account b via IAM roles, STS, trust policies, and get item.

Compare AWS Secrets Manager and Systems Manager Parameter Store for sensitive data in CI/CD pipelines; use Parameter Store for basic security or Secrets Manager for rotation and cross-account sharing.

Understand the difference between IAM user and IAM role, highlighting long-term identities for users and short-term identities that services or applications can assume to access AWS resources like EC2.