Deployment Strategies

Blue-green deployment maintains two identical environments: blue (current) and green (new). Traffic switches from blue to green in one step.

Process:

1. Deploy new version to green environment
2. Run tests against green
3. Switch load balancer to route all traffic to green
4. Monitor for issues
5. If issues occur, switch back to blue

Implementation with ALB:

# CloudFormation: create green target group
GreenTargetGroup:
  Type: AWS::ElasticLoadBalancingV2::TargetGroup
  Properties:
    HealthCheckPath: /health
    Port: 80
    VpcId: !Ref VPC

# Switch traffic by updating listener rule
ListenerRule:
  Type: AWS::ElasticLoadBalancingV2::ListenerRule
  Properties:
    Actions:
      - Type: forward
        TargetGroupArn: !Ref GreenTargetGroup

Advantages: instant rollback (switch back to blue), zero downtime, full testing of new version before traffic switch. Disadvantages: requires double infrastructure during deployment, more complex setup.

Canary deployment gradually shifts traffic from old to new version. Start with a small percentage (1-5%), monitor for issues, then increase gradually.

Process:

1. Deploy new version alongside old version
2. Route 5% of traffic to new version
3. Monitor error rates, latency, and metrics
4. Gradually increase: 5% -> 25% -> 50% -> 100%
5. Roll back if any metric degrades

Implementation with CodeDeploy:

DeploymentConfig:
  Type: AWS::CodeDeploy::DeploymentConfig
  Properties:
    DeploymentConfigName: Canary10Percent5Minutes
    MinimumHealthyHosts:
      Type: FLEET_PERCENT
      Value: 90
    TrafficRoutingConfig:
      Type: CanarySwitchTimeBased
      TimeBasedCanary:
        CanaryPercentage: 10
        CanaryInterval: 5

Canary with Lambda alias traffic shifting:

TrafficShiftingConfig:
  Type: Canary10Percent5Minutes
  LambdaFunctionVerifyStepNotifications: true

Advantages: reduced risk with gradual exposure, real-time monitoring of production traffic. Disadvantages: longer deployment time, requires sophisticated monitoring.

Rolling deployment replaces instances one at a time or in batches. At any point, a mix of old and new versions serves traffic.

Process:

1. Take one instance out of service
2. Deploy new version to that instance
3. Run health checks
4. Return instance to service
5. Repeat for next instance

CodeDeploy rolling deployment with ECS:

DeploymentConfig:
  Type: AWS::CodeDeploy::DeploymentConfig
  Properties:
    DeploymentConfigName: CodeDeployDefault.ECSCanary10Percent5Minutes
    ComputePlatform: ECS
    TrafficRoutingConfig:
      Type: CanarySwitchTimeBased
      TimeBasedCanary:
        CanaryPercentage: 10
        CanaryInterval: 5

Instance replacement strategy: CodeDeploy can use linear, canary, or all-at-once deployment configurations. For EC2, it terminates old instances and launches new ones. For ECS, it updates the task definition and shifts traffic.

Advantages: simple, no extra infrastructure needed, continuous availability. Disadvantages: slower than blue-green, mixed versions during deployment, harder to rollback quickly.

AWS CodeDeploy manages deployments across EC2, ECS, Lambda, and on-premises.

Application: The compute platform (EC2/On-Premises, ECS, Lambda).

Deployment group: Targets for deployment (EC2 instances by tag, ECS services, Lambda aliases). Each group has its own deployment configuration and hooks.

Deployment configuration: Defines deployment behavior (canary, linear, all-at-once).

Lifecycle hooks run scripts at specific points:

• BeforeInstall: pre-deployment setup
• AfterInstall: post-install configuration
• BeforeApplicationStart: start dependencies
• AfterApplicationStart: verify application started
• BeforeAllowTraffic: pre-traffic validation
• AfterAllowTraffic: post-traffic validation

DeploymentGroup:
  Type: AWS::CodeDeploy::DeploymentGroup
  Properties:
    ApplicationName: !Ref CodeDeployApp
    DeploymentConfigName: CodeDeployDefault.LinearAllAtOnceIn10Minutes
    AutoRollbackConfiguration:
      Enabled: true
      Events: [DEPLOYMENT_FAILURE, DEPLOYMENT_STOP_ON_REQUEST]
    AlarmConfiguration:
      Alarms:
        - !Ref ErrorAlarm

Auto rollback triggers automatically revert if alarms fire during deployment. Configure CloudWatch alarms for error rates, latency, or custom metrics.