devops（DevOps 部署）

[IMPORTANT] Use TaskCreate to break ALL work into small tasks BEFORE starting — including tasks for each file read. This prevents context loss from long files. For simple tasks, AI MUST ATTENTION ask user whether to skip.

Quick Summary

Goal: Deploy and manage cloud infrastructure across Cloudflare (Workers, R2, D1), Docker containers, and Google Cloud Platform.

Workflow:

1. Platform Selection — Choose Cloudflare (edge/low-latency), Docker (containers/microservices), or GCP (enterprise/K8s)
2. Project Setup — Initialize with Wrangler CLI, Dockerfile, or gcloud CLI
3. Local Development — Test locally before deploying
4. Deploy & Verify — Deploy to target platform with health checks

Key Rules:

• Run containers as non-root user; scan images for vulnerabilities
• Use multi-stage Docker builds to minimize image size
• Store secrets in environment variables, never in code
• Use R2 over S3 when zero egress cost matters

Be skeptical. Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence percentages (Idea should be more than 80%).

DevOps Skill

Comprehensive guide for deploying and managing cloud infrastructure across Cloudflare edge platform, Docker containerization, and Google Cloud Platform.

When to Use This Skill

Use this skill when:

• Deploying serverless applications to Cloudflare Workers
• Containerizing applications with Docker
• Managing Google Cloud infrastructure with gcloud CLI
• Setting up CI/CD pipelines across platforms
• Optimizing cloud infrastructure costs
• Implementing multi-region deployments
• Building edge-first architectures
• Managing container orchestration with Kubernetes
• Configuring cloud storage solutions (R2, Cloud Storage)
• Automating infrastructure with scripts and IaC

Platform Selection Guide

When to Use Cloudflare

Best For:

• Edge-first applications with global distribution
• Ultra-low latency requirements (<50ms)
• Static sites with serverless functions
• Zero egress cost scenarios (R2 storage)
• WebSocket/real-time applications (Durable Objects)
• AI/ML at the edge (Workers AI)

Key Products:

• Workers (serverless functions)
• R2 (object storage, S3-compatible)
• D1 (SQLite database with global replication)
• KV (key-value store)
• Pages (static hosting + functions)
• Durable Objects (stateful compute)
• Browser Rendering (headless browser automation)

Cost Profile: Pay-per-request, generous free tier, zero egress fees

When to Use Docker

Best For:

• Local development consistency
• Microservices architectures
• Multi-language stack applications
• Traditional VPS/VM deployments
• Kubernetes orchestration
• CI/CD build environments
• Database containerization (dev/test)

Key Capabilities:

• Application isolation and portability
• Multi-stage builds for optimization
• Docker Compose for multi-container apps
• Volume management for data persistence
• Network configuration and service discovery
• Cross-platform compatibility (amd64, arm64)

Cost Profile: Infrastructure cost only (compute + storage)

When to Use Google Cloud

Best For:

• Enterprise-scale applications
• Data analytics and ML pipelines (BigQuery, Vertex AI)
• Hybrid/multi-cloud deployments
• Kubernetes at scale (GKE)
• Managed databases (Cloud SQL, Firestore, Spanner)
• Complex IAM and compliance requirements

Key Services:

• Compute Engine (VMs)
• GKE (managed Kubernetes)
• Cloud Run (containerized serverless)
• App Engine (PaaS)
• Cloud Storage (object storage)
• Cloud SQL (managed databases)

Cost Profile: Varied pricing, sustained use discounts, committed use contracts

Quick Start

Cloudflare Workers

# Install Wrangler CLI
npm install -g wrangler

# Create and deploy Worker
wrangler init my-worker
cd my-worker
wrangler deploy

See: references/cloudflare-workers-basics.md

Docker Container

# Create Dockerfile
cat > Dockerfile <<EOF
FROM node:20-alpine
WORKDIR /app
COPY package*.json ./
RUN npm ci --production
COPY . .
EXPOSE 3000
CMD ["node", "server.js"]
EOF

# Build and run
docker build -t myapp .
docker run -p 3000:3000 myapp

See: references/docker-basics.md

Google Cloud Deployment

# Install and authenticate
curl https://sdk.cloud.google.com | bash
gcloud init
gcloud auth login

# Deploy to Cloud Run
gcloud run deploy my-service \
  --image gcr.io/project/image \
  --region us-central1

See: references/gcloud-platform.md

Reference Navigation

Cloudflare Platform

• cloudflare-platform.md - Edge computing overview, key components
• cloudflare-workers-basics.md - Getting started, handler types, basic patterns
• cloudflare-workers-advanced.md - Advanced patterns, performance, optimization
• cloudflare-workers-apis.md - Runtime APIs, bindings, integrations
• cloudflare-r2-storage.md - R2 object storage, S3 compatibility, best practices
• cloudflare-d1-kv.md - D1 SQLite database, KV store, use cases
• browser-rendering.md - Puppeteer/Playwright automation on Cloudflare

Docker Containerization

• docker-basics.md - Core concepts, Dockerfile, images, containers
• docker-compose.md - Multi-container apps, networking, volumes

Google Cloud Platform

• gcloud-platform.md - GCP overview, gcloud CLI, authentication
• gcloud-services.md - Compute Engine, GKE, Cloud Run, App Engine

Python Utilities

• scripts/cloudflare-deploy.py - Automate Cloudflare Worker deployments
• scripts/docker-optimize.py - Analyze and optimize Dockerfiles

Common Workflows

Edge + Container Hybrid

# Cloudflare Workers (API Gateway)
# -> Docker containers on Cloud Run (Backend Services)
# -> R2 (Object Storage)

# Benefits:
# - Edge caching and routing
# - Containerized business logic
# - Global distribution

Multi-Stage Docker Build

# Build stage
FROM node:20-alpine AS build
WORKDIR /app
COPY package*.json ./
RUN npm ci
COPY . .
RUN npm run build

# Production stage
FROM node:20-alpine
WORKDIR /app
COPY --from=build /app/dist ./dist
COPY --from=build /app/node_modules ./node_modules
USER node
CMD ["node", "dist/server.js"]

CI/CD Pipeline Pattern

# 1. Build: Docker multi-stage build
# 2. Test: Run tests in container
# 3. Push: Push to registry (GCR, Docker Hub)
# 4. Deploy: Deploy to Cloudflare Workers / Cloud Run
# 5. Verify: Health checks and smoke tests

Best Practices

Security

• Run containers as non-root user
• Use service account impersonation (GCP)
• Store secrets in environment variables, not code
• Scan images for vulnerabilities (Docker Scout)
• Use API tokens with minimal permissions

Performance

• Multi-stage Docker builds to reduce image size
• Edge caching with Cloudflare KV
• Use R2 for zero egress cost storage
• Implement health checks for containers
• Set appropriate timeouts and resource limits

Cost Optimization

• Use Cloudflare R2 instead of S3 for large egress
• Implement caching strategies (edge + KV)
• Right-size container resources
• Use sustained use discounts (GCP)
• Monitor usage with cloud provider dashboards

Development

• Use Docker Compose for local development
• Wrangler dev for local Worker testing
• Named gcloud configurations for multi-environment
• Version control infrastructure code
• Implement automated testing in CI/CD

Decision Matrix

Resources

• Cloudflare Docs: https://developers.cloudflare.com
• Docker Docs: https://docs.docker.com
• GCP Docs: https://cloud.google.com/docs
• Wrangler CLI: https://developers.cloudflare.com/workers/wrangler/
• gcloud CLI: https://cloud.google.com/sdk/gcloud

Implementation Checklist

Cloudflare Workers

• Install Wrangler CLI
• Create Worker project
• Configure wrangler.toml (bindings, routes)
• Test locally with wrangler dev
• Deploy with wrangler deploy

Docker

• Write Dockerfile with multi-stage builds
• Create.dockerignore file
• Test build locally
• Push to registry
• Deploy to target platform

Google Cloud

• Install gcloud CLI
• Authenticate with service account
• Create project and enable APIs
• Configure IAM permissions
• Deploy and monitor resources

Related

• databases
• api-design

Closing Reminders

• MANDATORY IMPORTANT MUST ATTENTION break work into small todo tasks using TaskCreate BEFORE starting
• MANDATORY IMPORTANT MUST ATTENTION search codebase for 3+ similar patterns before creating new code
• MANDATORY IMPORTANT MUST ATTENTION cite file:line evidence for every claim (confidence >80% to act)
• MANDATORY IMPORTANT MUST ATTENTION add a final review todo task to verify work quality