Entry questions

A repository at /home/interview/repo contains a deployment pipeline defined in .github/workflows/deploy.yml. Currently, the pipeline deploys to both Staging and Production environments automatically on every push. This is risky because unstable code could break the Production environment immediately.

A repository at /home/interview/repo contains a deployment workflow that occasionally fails due to transient issues. A deployment step running ./scripts/flaky-deploy.sh fails on first attempt but succeeds on retry. The workflow needs automatic retry configuration to handle transient failures. A starter workflow file has been created at .github/workflows/deploy.yml with the basic structure.

A repository at /home/interview/repo contains a deployment workflow that occasionally hangs due to network issues or unresponsive external services. Jobs must be automatically terminated if they exceed a time limit to prevent resource waste. A starter workflow file has been created at .github/workflows/deploy.yml with the basic structure.

A security audit has flagged the IAM user app-deployer for having AdministratorAccess policy with far more permissions than needed. The user only needs access to:

• S3: Read (GetObject) and Add objects (PutObject), list buckets (ListBucket).
• CloudWatch Logs: Create log groups (CreateLogGroup), log streams (CreateLogStream), and put log events (PutLogEvents).

Your team needs a simple Lambda function for a greeting microservice.

You need to set up a new administrative user account for regular use instead of using the root account.

Your team needs an EC2 instance to manage IAM resources programmatically. To follow AWS security best practices, you should use an IAM role instead of embedding credentials.

Your infrastructure has two application endpoints requiring distinct monitoring strategies through Route 53 health checks, each with different protocols, ports, health check intervals, and failure thresholds.

Your team is launching a new application and needs DNS configured in Route 53 for the domain example.com.

A basic landing page needs to be delivered via an EC2 instance accessible over HTTP with its own security group.

You have a container named cpu_test running from the image myapp:cpu that performs CPU-intensive computations. When monitoring with docker stats, the container regularly spikes to 150-200% CPU usage, consuming multiple CPU cores and impacting other services on the host.

You have a container image myapp:dualfix built from /home/interview/Dockerfile that fails during execution.

You have a Dockerfile at /home/interview/Dockerfile that needs to be built for multiple architectures to support different hardware platforms. Currently, building with docker build only creates images for the host architecture.

You need to create a Docker volume named myvol and demonstrate cross-platform consistency - that data written from one container is readable by a different container using a different base image.

A container image myapp:logapp continuously writes logs to stdout. Without log rotation configured, Docker's log files grow unbounded, consuming disk space and potentially filling up the filesystem.

You have multiple work-in-progress states saved in your Git stash stack and need to retrieve code from an older snapshot.

You are working on the main branch and need to update config.json with changes that currently exist only on the feature-settings branch, without switching your current context.

You have a Git repository at /home/interview/repo on the main branch. A commit on the feature branch contains a bug fix you need on main, but you don't want to merge the entire feature branch.

You have a Git repository at /home/interview/repo where the remote origin is currently configured with an HTTPS URL (https://github.com/user/repo.git).

You have a Git repository at /home/interview/repo where you've completed work on the feature-login branch. The feature is ready to be integrated into the main branch.

You have a Git repository at /home/interview/repo where your feature-refactor branch has four commits in the wrong logical order. The commits should be reversed (oldest to newest becomes newest to oldest) to make more sense for code review. You need to rebase this branch onto the latest main branch while reversing the commit order.

You have a Git repository at /home/interview/repo where your feature-payment branch is 32 commits behind main. You need to rebase the feature branch onto the latest main and resolve conflicts that occur during the process.

A Git repository at /home/interview/repo has a feature branch feature-login with several commits, but it was created from the wrong base branch.

You have created a commit locally that contains incorrect changes and you want to completely erase it from history.

A deployment pipeline is experiencing slow build times because it performs full history clones.

You have a Git repository at /home/interview/repo where you modified three files: app.js, style.css, and config.json.

You have a Git repository at /home/interview/repo with three recent commits on the main branch that were premature and need to be reorganized.

A Git repository at /home/interview/repo has a feature-api branch that was created from origin/main some time ago. The branch has your commits plus merge commits from pulling updates, making it difficult to identify which commits are unique to your branch.

You need to create a new namespace for deploying applications.

An application requires persistent storage that can be resized without downtime.

A Job manifest at /home/interview/fail-job.yaml is failing with a non-zero exit status. The Job has limited retries configured.

An application requires multiple configuration values mounted as files.

A monitoring application needs read-only access to pods in the demo namespace.

You need to deploy a web application that only receives traffic when it's ready to serve requests.

An application needs guaranteed resources and limits to prevent overconsumption.

During peak traffic hours, a background data-processing job that is started by user devops is consuming too much CPU and slowing down user-facing services. Stopping it isn't an option because it would interrupt ongoing workflows, but you need to limit its CPU usage temporarily.

The web server has become sluggish and users are experiencing timeout errors. Multiple worker processes are running under the nginx user, and one of them is consuming excessive CPU resources, causing the entire application to slow down.

During an incident investigation, you pulled a massive log export from /var/log/app/access.log that's several gigabytes in size. Your analysis tools and editors can't handle the entire file at once.

The /var directory contains logs from multiple applications, and cleanup planning is needed. Some applications create their own subdirectories with nested log files.

Users are complaining about slow file access. System metrics show high disk utilization.

An application /home/interview/server.sh fails to start.

The /tmp directory has accumulated numerous leftover folders from previous application runs and temporary scripts. Many of these directories are now empty and can be safely removed.

Before performing a system-wide database schema migration, you've been asked to ensure that all existing .db files are safely backed up. These files may be scattered across multiple subdirectories, and simply renaming them isn't enough. You must create backup copies with the .db.bak extension, preserving directory structure and permissions.

Multiple applications write logs under /var/log, and you need to quickly check if any recent errors have been recorded.

You have a log file /tmp/app_combined.log with entries from multiple servers in mixed order. Each line contains: timestamp, hostname, and action.

A server is experiencing latency spikes and intermittent timeouts. CPU and memory appear normal, suggesting a disk I/O bottleneck from a runaway process.

The /var/log/messages file has been growing unusually fast, filling up disk space within hours.

System resources are being consumed by an unusually large process tree. You need to identify the parent process with the most children and document its hierarchy.

A Linux server is under review for stability and recent performance evaluation.

Users are reporting random timeouts. You need to determine whether packet loss occurs on the local network, upstream gateway, or public internet.

A security audit found several unauthorized applications listening on high ports (range 8000-9000).

Network latency issues were reported due to opening too many TCP connections.

You need to reach a remote subnet that isn't covered by the default route, but you don't want to make any permanent configuration changes.

Your monitoring alerts indicate that DNS resolution might be inconsistent across servers, and you need to confirm both IPv4 and IPv6 records for a given domain.