Docker Scenario-Based Interview Questions and Detailed Answers (2025)

Preparing for a Docker interview in 2025? Scenario-based questions test your practical problem-solving skills and real-world experience with Docker. This blog covers 10 complex scenarios with detailed solutions, from debugging to production deployments.

1. Scenario: Debugging a Crashing Container

Problem:
A container running a Python Flask app crashes randomly with exit code 137.

Solution Steps:

1. Check Logs:

   docker logs <container_id>
   

   • Look for Out of Memory (OOM) errors.

2. Inspect Resource Limits:

   docker inspect <container_id> | grep -i "memory"
   

   • If no limits are set, the host might kill the container under memory pressure.

3. Fix:

   • Limit Memory:

     docker run -d --memory="512m" --name flask_app flask:latest
     

   • Optimize App: Reduce memory leaks (e.g., unbounded caches).

2. Scenario: Docker Build is Too Slow

Problem:
Building a Docker image takes 15 minutes due to large dependencies.

Solution Steps:

1. Optimize Dockerfile:

   • Use multi-stage builds to discard build tools:

     FROM python:3.9 as builder
     COPY requirements.txt .
     RUN pip install --user -r requirements.txt
     
     FROM python:3.9-slim
     COPY --from=builder /root/.local /root/.local
     COPY . .
     

2. Leverage Build Cache:

   • Order commands from least to most frequently changed:

     COPY requirements.txt .  # Rarely changes → cached
     RUN pip install -r requirements.txt
     COPY . .                 # Frequently changes → last
     

3. Use .dockerignore:

   node_modules/
   .git/
   

3. Scenario: Container Can’t Connect to Database

Problem:
A containerized Node.js app fails to connect to a PostgreSQL DB with ECONNREFUSED.

Solution Steps:

1. Verify Network:

   docker network ls
   docker inspect <network_id> | grep -A 5 "Containers"
   

   • Ensure both containers are on the same network.

2. Test Connectivity:

   docker exec -it node_app ping db
   

   • If ping fails, check DNS resolution.

3. Fix:

   • Create a shared network:

     docker network create app_net
     docker run -d --net=app_net --name db postgres
     docker run -d --net=app_net --name node_app node:latest
     

4. Scenario: Persistent Data Loss After Container Restart

Problem:
A MySQL container loses data when restarted.

Solution Steps:

1. Check Volume Usage:

   docker inspect mysql_container | grep -A 10 "Mounts"
   

   • If no volume is mounted, data is stored in the container’s writable layer (ephemeral).

2. Fix:

   • Use a Named Volume:

     docker volume create mysql_data
     docker run -d -v mysql_data:/var/lib/mysql --name mysql mysql:latest
     

   • Backup Volume:

     docker run --rm -v mysql_data:/data -v $(pwd):/backup busybox tar cvf /backup/mysql_backup.tar /data
     

5. Scenario: Docker Host Running Out of Disk Space

Problem:
The Docker host throws No space left on device errors.

Solution Steps:

1. Identify Large Files:

   docker system df  # Show disk usage
   docker images --filter "dangling=true" -q | xargs docker rmi  # Remove dangling images
   

2. Prune Unused Objects:

   docker system prune -a --volumes
   

3. Limit Log Size:

   {
     "log-driver": "json-file",
     "log-opts": {
       "max-size": "10m",
       "max-file": "3"
     }
   }
   

   (Configure in /etc/docker/daemon.json)

6. Scenario: Securing a Container Running as Root

Problem:
A container runs as root, posing security risks.

Solution Steps:

1. Run as Non-Root:

   FROM alpine
   RUN adduser -D appuser
   USER appuser
   

2. Drop Capabilities:

   docker run --cap-drop=ALL --cap-add=NET_BIND_SERVICE nginx
   

3. Read-Only Filesystem:

   docker run --read-only nginx
   

7. Scenario: Deploying a Multi-Service App with Docker Compose

Problem:
Need to deploy a web app (Nginx), API (Node.js), and DB (PostgreSQL) with dependencies.

Solution Steps:

1. Write docker-compose.yml:

   version: "3.8"
   services:
     web:
       image: nginx
       ports: ["80:80"]
     api:
       build: ./api
       depends_on: ["db"]
     db:
       image: postgres
       volumes: ["pg_data:/var/lib/postgresql/data"]
   volumes:
     pg_data:
   

2. Deploy:

   docker-compose up -d
   

8. Scenario: Zero-Downtime Deployment with Docker Swarm

Problem:
Update a service without downtime.

Solution Steps:

1. Rolling Update:

   docker service update \
     --image myapp:v2 \
     --update-parallelism 2 \
     --update-delay 10s \
     myapp
   

2. Rollback on Failure:

   docker service update --rollback myapp
   

9. Scenario: Debugging Slow Container Startup

Problem:
A container takes 2 minutes to start.

Solution Steps:

1. Profile Startup:

   time docker run myapp
   

2. Check Entrypoint:

   docker inspect myapp | grep -A 5 "Entrypoint"
   

   • Optimize slow init scripts.

3. Use --init for Zombie Processes:

   docker run --init myapp
   

10. Scenario: Sharing Data Between Containers

Problem:
Two containers need access to the same configuration file.

Solution Steps:

1. Use a Shared Volume:

   docker volume create config_vol
   docker run -d -v config_vol:/config --name container1 alpine
   docker run -d -v config_vol:/config --name container2 alpine
   

2. Alternative: Bind Mount:

   docker run -d -v /host/path:/config --name container1 alpine
   

Final Tips for Scenario-Based Interviews

1. Ask Clarifying Questions:
   • "Is the container running in Swarm/Kubernetes?"
   • "What’s the error message?"
2. Explain Trade-offs:
   • "Bind mounts are easier for dev, but volumes are better for prod."
3. Show Debugging Steps:
   • "First, I’d check docker logs, then docker inspect..."

Need more scenarios? Drop a comment! 🚀

Practice Challenge:

• Simulate a memory leak in a container and debug it using docker stats and exec.
• Deploy a 3-tier app (frontend, backend, DB) using Compose.

Good luck with your interview! 🐳