deliverables.md

Project Validation and Troubleshooting Guide

1. Component 3: Async Communication (Kafka)

Validation Steps:

Check Kafka Topics:

kubectl exec -it kafka-68599c8986-8gcdj -- /opt/bitnami/kafka/bin/kafka-topics.sh \
  --list --bootstrap-server localhost:9092

✅ Should show sensor_topic

Verify Kafka Messages:

kubectl exec -it kafka-68599c8986-8gcdj -- /opt/bitnami/kafka/bin/kafka-console-consumer.sh \
  --topic sensor_topic --from-beginning --bootstrap-server localhost:9092

✅ Should show continuous JSON sensor data

Check Server Kafka Logs:

kubectl logs deployment/server | grep "Kafka received"

✅ Should show received sensor data

---

2. Component 4: Leader Election & Consensus (Raft)

Validation Steps:

Confirm Leader Election:

kubectl logs deployment/server | grep "Elected leader"

✅ Example Output: Elected leader at 10.244.0.5:50051 (Term 1)

Test Leader Redirection:

kubectl logs deployment/client | grep "Redirecting"

✅ Clients should redirect to valid leader IPs

Kill Leader Pod:

kubectl delete pod <leader-pod-name>

✅ New leader should be elected automatically (check server logs)

---

3. Component 5: Auto-Scaling (Kubernetes HPA)

Validation Steps:

Check HPA Status:

kubectl get hpa

✅ Verify CPU utilization triggers scaling

Generate Load:

kubectl exec -it client-56ccb5499d-db4nn -- python3 scripts/load_test.py

✅ Load should trigger scaling

Observe Scaling:

watch kubectl get pods

✅ Server pods should scale up when load increases

---

4. Component 6: Resource Management

Validation Steps:

Verify Resource Limits:

kubectl describe pod server-74dfddc7c8-c2qlv | grep -A 5 "Resources"

✅ Should show CPU/memory limits from server-deployment.yaml

Check Priority Classes:

kubectl get priorityclass

✅ Verify high-priority class exists

---

5. Non-Functional Requirements

Validation Steps:

Metrics Endpoint:

kubectl port-forward deployment/server 8000:8000

✅ Open http://localhost:8000/metrics - should show grpc_requests_total

Latency Test:

kubectl logs deployment/client | grep "Optimized settings received"

✅ Check for consistent response times (<1s)

Chaos Test:

kubectl delete pod kafka-68599c8986-8gcdj  # Kill Kafka

✅ Clients should buffer messages until Kafka recovers

---

6. Final Deliverables Checklist

Requirement	Validation Method
System Design Diagram	Confirm diagram exists in project docs
Demo Video (1-2 mins)	Verify video shows all components working
Performance Metrics	Check Prometheus/Grafana dashboard
Auto-Scaling Proof	kubectl describe hpa + load test screenshots
Code Quality	All pods show 0 restarts in kubectl get pods

---

Troubleshooting Tips

No Leader Elected:

• Check Raft logic in raft.py
• Ensure POD_IP is injected in server pods:

  kubectl describe pod <server-pod-name>

Kafka Connection Issues:

• Verify Kafka service DNS name:

  kubectl get svc kafka

• Check Zookeeper logs:

  kubectl logs zookeeper-5968d887b9-crpbx

gRPC Errors:

• Verify server-service:50051 exists:

  kubectl get svc server-service

---

✅ If all these checks pass, your system is 100% operational and meets all project requirements! 🚀