Cloud & DevOps Nodes

I nodi Cloud & DevOps permettono di automatizzare pipeline CI/CD, build Docker, deploy su cloud e gestione infrastruttura.

Docker Build

dockerBuild

Costruisce immagini Docker da Dockerfile o build context.

Configurazione

Campo	Tipo	Obbligatorio	Descrizione
`label`	text	Sì	Nome identificativo
`contextPath`	text	Sì	Path al build context
`dockerfile`	text	No	Path Dockerfile (default: Dockerfile)
`imageName`	text	Sì	Nome immagine
`tag`	text	Sì	Tag immagine
`buildArgs`	keyvalue	No	Build arguments
`target`	text	No	Target stage per multi-stage
`platform`	select	No	Piattaforma target
`push`	boolean	No	Push a registry dopo build
`registry`	text	Per push	Registry URL

Workflow Esempio: CI/CD Pipeline

Diagramma di flusso

Configurazione build

type: dockerBuild
contextPath: "/app"
dockerfile: "Dockerfile.prod"
imageName: "myapp"
tag: "{{trigger.data.commit.sha | slice: 0, 7}}"
buildArgs:
  NODE_ENV: "production"
  VERSION: "{{trigger.data.commit.sha}}"
target: "production"
platform: "linux/amd64"
push: true
registry: "registry.example.com"
registryUser: "{{secrets.REGISTRY_USER}}"
registryPassword: "{{secrets.REGISTRY_PASS}}"
label: "Build Production Image"

Output

{
  "imageName": "registry.example.com/myapp",
  "tag": "abc1234",
  "fullName": "registry.example.com/myapp:abc1234",
  "digest": "sha256:...",
  "size": "125MB",
  "buildTime": 45000,
  "pushed": true,
  "layers": 12,
  "buildLog": "Step 1/12: FROM node:18-alpine..."
}

Multi-stage Build

Esempio di Dockerfile ottimizzato per produzione.

Dockerfile multi-stage

# Stage 1: Build
FROM node:18-alpine AS builder
WORKDIR /app
COPY package*.json ./
RUN npm ci --only=production
COPY . .
RUN npm run build

# Stage 2: Production
FROM node:18-alpine AS production
WORKDIR /app
COPY --from=builder /app/dist ./dist
COPY --from=builder /app/node_modules ./node_modules
EXPOSE 3000
CMD ["node", "dist/server.js"]

Configurazione per stage specifico

type: dockerBuild
contextPath: "/app"
target: "production"  # Usa solo lo stage production
imageName: "myapp"
tag: "latest"

Cloud Deploy

cloudDeploy

Deploy di applicazioni su cloud provider (AWS, GCP, Azure, DigitalOcean, etc.).

Configurazione

Campo	Tipo	Obbligatorio	Descrizione
`label`	text	Sì	Nome identificativo
`provider`	select	Sì	Cloud provider
`service`	select	Sì	Tipo servizio
`region`	text	Sì	Regione deploy
`credentials`	secret	Sì	Credenziali cloud
`config`	json	Sì	Configurazione specifica

Provider supportati

Provider	Servizi
`aws`	ECS, Lambda, EC2, S3, CloudFront
`gcp`	Cloud Run, GKE, Compute Engine, Cloud Functions
`azure`	App Service, AKS, Container Instances
`digitalocean`	App Platform, Droplets, Kubernetes
`vercel`	Static, Serverless
`fly`	Fly Machines

Workflow Esempio: Deploy AWS ECS

Diagramma di flusso

Configurazione AWS ECS

type: cloudDeploy
provider: aws
service: ecs
region: "eu-west-1"
credentials:
  accessKeyId: "{{secrets.AWS_ACCESS_KEY}}"
  secretAccessKey: "{{secrets.AWS_SECRET_KEY}}"
config:
  cluster: "production"
  service: "myapp-service"
  taskDefinition: "myapp-task"
  image: "{{dockerBuild.output.fullName}}"
  desiredCount: 3
  deploymentConfig:
    maximumPercent: 200
    minimumHealthyPercent: 50
  loadBalancer:
    targetGroupArn: "arn:aws:elasticloadbalancing:..."
    containerName: "myapp"
    containerPort: 3000
label: "Deploy ECS Production"

Output

{
  "provider": "aws",
  "service": "ecs",
  "deploymentId": "ecs-svc/abc123",
  "status": "success",
  "taskDefinition": "myapp-task:42",
  "desiredCount": 3,
  "runningCount": 3,
  "pendingCount": 0,
  "events": [
    {
      "timestamp": "2024-01-15T10:35:00Z",
      "message": "service myapp-service has reached a steady state."
    }
  ],
  "deployedAt": "2024-01-15T10:30:00Z"
}

Deploy Kubernetes

Deploy su cluster Kubernetes.

Workflow Esempio: K8s Rolling Update

Diagramma di flusso

Configurazione Kubernetes

type: cloudDeploy
provider: kubernetes
config:
  kubeconfig: "{{secrets.KUBECONFIG}}"
  namespace: "production"
  deployment: "myapp"
  container: "main"
  image: "{{dockerBuild.output.fullName}}"
  strategy:
    type: RollingUpdate
    maxUnavailable: 1
    maxSurge: 1
  replicas: 3
  resources:
    requests:
      cpu: "100m"
      memory: "256Mi"
    limits:
      cpu: "500m"
      memory: "512Mi"
label: "K8s Deploy"

Deploy Serverless

Deploy di funzioni serverless.

Workflow Esempio: AWS Lambda

Diagramma di flusso

Configurazione Lambda

type: cloudDeploy
provider: aws
service: lambda
region: "eu-west-1"
credentials:
  accessKeyId: "{{secrets.AWS_ACCESS_KEY}}"
  secretAccessKey: "{{secrets.AWS_SECRET_KEY}}"
config:
  functionName: "myFunction"
  runtime: "nodejs18.x"
  handler: "index.handler"
  zipFile: "/tmp/function.zip"
  memorySize: 256
  timeout: 30
  environment:
    NODE_ENV: "production"
    DB_URL: "{{secrets.DB_URL}}"
label: "Deploy Lambda"

CI/CD Pipeline Completa

Pipeline completa con test, build, deploy staging e production.

Diagramma di flusso

Blue-Green Deployment

Strategia di deploy con zero downtime.

Diagramma di flusso

Workflow Blue-Green

Diagramma di flusso

Canary Deployment

Deploy graduale con percentuale crescente di traffico.

Diagramma di flusso

Best Practices

Immagini Docker

Usa multi-stage build per immagini più piccole
Tag immutabili (commit SHA) oltre a latest
Scan vulnerabilità prima del push
Layer caching per build veloci

Deploy sicuri

Sempre health check dopo deploy
Rollback automatico in caso di errori
Deploy progressivi (canary/blue-green)
Feature flags per rilasci controllati

Secrets management

Mai credenziali in immagini Docker
Usa secrets manager (AWS Secrets Manager, Vault)
Rotazione automatica delle credenziali
Least privilege per credenziali CI/CD

Monitoring deploy

Metriche chiave: error rate, latency, throughput
Alert automatici su anomalie post-deploy
Logging centralizzato per debug
Trace distribuito per problemi complessi