Self:

In my previous position as DevOps/SRE engineer, my day-to-day tasks were a mix of both infrastructure operational activities and deployments and audit and analysis of system performance and logs, root cause analysis, Coordinate Migration tasks etc

•  kubernet cluster management like, node scaling, resource limiting, troubleshoot pod failure, network issues etc

Kubernetes:

CrashLoopBackOff - Container fails repeatedly on startup

🔹 Check logs: kubectl logs <pod>

🔹 Check readiness/liveness probes

🔹 Validate CMD/ENTRYPOINT in Dockerfile

ImagePullBackOff / ErrImagePull - Image not found or authentication failed

🔹 Check image name/tag

🔹 Verify registry credentials / secret

🔹 Run kubectl describe pod <pod>

RunContainerError - Container failed to start

🔹 Check Dockerfile CMD/ENTRYPOINT

🔹 Ensure image isn't corrupted

CreateContainerConfigError - Pod spec config issue

🔹 Check ConfigMap/Secret references

🔹 Run: kubectl describe pod <pod>

Pending - Pod can’t be scheduled

🔹 No matching node (due to resource limits, taints)

🔹 Check with: kubectl describe pod <pod> and kubectl get nodes

NodeNotReady - Node is not available

🔹 Check kubectl get nodes

🔹 Restart kubelet or fix node issues

Back-off restarting failed container - App inside container is crashing

🔹 Check logs

🔹 Debug locally or update health checks

OOMKilled - Out Of Memory (container exceeded limits)

🔹 Increase memory limit in YAML

🔹 Optimize application memory usage

Evicted - Pod killed due to low node resources

🔹 Add more nodes

🔹 Tweak pod resource limits/requests

Unauthorized / Forbidden - RBAC issues

🔹 Review RBAC roles and bindings

🔹 Check service account permissions

dns: service not found - DNS resolution failed

🔹 Check Service/Pod name

🔹 Check CoreDNS status: kubectl get pods -n kube-system

no matches for kind - Typo or wrong API version in YAML

🔹 Use kubectl api-resources or kubectl explain <kind>

timeout connecting to service - Network or port issue

🔹 Validate service, selector, targetPort

🔹 Use kubectl port-forward to test

Secret not found Missing secret in namespace

🔹 Run kubectl get secrets

🔹 Check namespace with -n <namespace>

Error from server (NotFound) - Resource doesn't exist

🔹 Double-check the name and namespace

invalid character 'i' looking for beginning of value

Bad YAML or JSON input

🔹 Validate YAML syntax (e.g. with yamllint)

✅ Troubleshooting Tools

kubectl describe pod <pod> - Deep inspection of pod issues

kubectl logs <pod> - View logs from containers

kubectl get events --sort-by='.metadata.creationTimestamp' - View cluster events

kubectl top pod - View resource usage

kubectl port-forward - Access services locally

kubectl get all -n <namespace> - Overview of resources in namespace

Docker

1. Build and Run Containers

🔹 Task: docker build -t myapp .

Error:failed to solve: failed to compute cache key: not found

Cause: Missing Dockerfile or bad context
Fix: Make sure the Dockerfile exists in the directory.

Use: docker build -f path/to/Dockerfile .

🔹 Task: docker run -d -p 8080:80 myapp

Error: port is already allocated

Cause: Host port 8080 already in use
Fix: Use a different port: docker run -d -p 8081:80 myapp

Or free the port: sudo lsof -i :8080

📦 2. Image & Container Management

🔹 Task: docker pull <image>

Error: repository does not exist or may require 'docker login'

Cause: Private repo or typo in name
Fix: Check image name spelling

Use: docker login if private registry

🔹 Task: docker rmi <image>

Error: image is being used by stopped container

Fix:Remove container first: docker rm <container_id>

Then: docker rmi <image_id>

🛠️ 3. Working with Volumes and Mounts

🔹 Error:

Mounts denied: file not shared from host

Cause: MacOS/Windows file sharing is restricted
Fix:Enable folder sharing in Docker Desktop → Settings → Resources → File Sharing

🔹 Error:

permission denied when writing inside mounted volume

Cause: Host file system permissions
Fix:Use chown inside Dockerfile

Or run container with same UID as host user

📶 4. Network and Connectivity

🔹 Error: curl: (6) Could not resolve host

Cause: DNS resolution failed inside container
Fix: Restart Docker: systemctl restart docker

Check Docker’s DNS settings (in /etc/docker/daemon.json)

🔹 Error: connection refused

Cause: Target service in another container is not reachable
Fix: Ensure containers are on same network

Use Docker network aliases

🧼 5. Cleanup & Pruning

🔹 Task: docker system prune -a

Error: Error response from daemon: conflict: unable to remove repository

Fix: Remove dependent containers/images first

Use: docker container prune

docker image prune -a

🔍 6. Logs, Exec, and Debug

🔹 Task: docker logs <container>

Error: no such container

Fix: List containers: docker ps -a

Check name or ID is correct

🔹 Task: docker exec -it <container> bash

Error: exec failed: container_linux.go: no such file or directory

Cause: Image doesn’t have bash
Fix:Try: docker exec -it <container> sh

🧰 General Troubleshooting Commands

docker ps -a  =  List all containers

docker logs <id> = View container logs

docker inspect <id> = View detailed metadata

docker events = Monitor Docker daemon events

docker-compose logs = Check logs in multi-container setup

SHELL SCRIPTING

AWS

LINUX