Container as a Service

Run and scale containerized workloads without managing Kubernetes clusters or virtual machines.

Overview

Container as a Service (CaaS) handles infrastructure so you can focus on your application:

• Scheduling and lifecycle management
• Scaling and auto-healing
• Load balancing and networking
• Monitoring, logging, and cost optimization
• CI/CD integration with GitHub Actions

---

Prerequisites

• Project dashboard or API access
• Container image (public or private registry)
• (Optional) Registry credentials for private images
• (Optional) Domain or subdomain for public endpoints
• (Optional) GitHub repository for automated deployments

---

Step 1: Create Container Instance

1. Navigate to Compute > Container as a Service
2. Click Create Container
3. Configure:

Setting	Description
Container Image	e.g., registry.example.com/app:latest
Registry Credentials	Required for private registries
Startup Command	Optional entrypoint override
Region/Zone	Deployment location
CPU and Memory	Resource allocation
Autoscaling	Min/max instances (e.g., 1-5)

4. Click Create

Deployment takes a few minutes. Status becomes Active when ready.

---

Step 2: Configure Runtime

Setting	Description
Ports and Protocols	HTTP, HTTPS, TCP exposure
Environment Variables	Credentials, API keys, configs
Scaling Rules	CPU, memory, or request rate thresholds
Container Lifetime	Scale-to-zero when idle
Secrets Management	Secure environment bindings

---

Step 3: Networking and Access

• Each container gets a unique endpoint (public or private)
• Map custom domain via CNAME record
• Configure TLS certificates (upload or auto-generate)
• Restrict to private subnets for internal systems

---

Step 4: CI/CD with GitHub Actions

Example Workflow

name: Deploy Container
on:
  push:
    branches: [ "main" ]

jobs:
  build-and-deploy:
    runs-on: ubuntu-latest
    steps:
      - name: Checkout code
        uses: actions/checkout@v4

      - name: Build Docker image
        run: docker build -t registry.example.com/myapp:${{ github.sha }} .

      - name: Log in to registry
        run: echo ${{ secrets.REGISTRY_PASSWORD }} | docker login registry.example.com -u ${{ secrets.REGISTRY_USER }} --password-stdin

      - name: Push image
        run: docker push registry.example.com/myapp:${{ github.sha }}

      - name: Deploy to CaaS
        run: |
          curl -X POST https://api.yourplatform.net/v1/containers/deploy \
          -H "Authorization: Bearer ${{ secrets.CAAS_API_KEY }}" \
          -d '{"image": "registry.example.com/myapp:'${{ github.sha }}'"}'

Extend with automated tests, Slack notifications, or blue-green releases.

---

Step 5: Monitoring and Logs

• Monitor status, CPU/memory usage, instance count via dashboard
• Stream logs via console or API:

caas logs <container-id>

• Integrate with ELK, Loki, or CloudWatch
• Set up alerts for latency or failed deploys

---

Step 6: Lifecycle Management

Action	Description
Start/Stop/Restart	Control containers anytime
Redeploy	Update with new image tag
Delete	Free resources from old instances
Rollback	Revert to previous stable image

---

Features Summary

Feature	Description
Managed Runtime	No server or cluster management
Registry Support	Public and private registries
Autoscaling	Demand-based scaling
Scale-to-Zero	Suspend idle containers
Custom Domains	Branded URL endpoints
CI/CD Integration	GitHub Actions automation
Secrets Management	Secure config injection
Monitoring and Logging	Centralized metrics and logs
Pay-as-you-go	Charged only while running

---

Troubleshooting

Issue	Cause	Solution
Container stuck in Pending	Quota exceeded or image pull issue	Check capacity and registry credentials
GitHub Actions deploy failed	Invalid API token	Verify CAAS_API_KEY secret
Logs not visible	Logging disabled	Enable log streaming
Domain not resolving	Incorrect DNS	Verify CNAME and propagation
Autoscaling not triggering	Thresholds too high	Lower triggers, verify monitoring