- First is global health dashboard
- Service Health Dashboard - health of AWS itself and its services
- All services, all regions - problem is, it's too big
- Personal health dashboard
- Event log for things that apply to you
- Click on bell to see
- Things like operational issues with service (lambda/EC2/ELB/etc), upgrades, VPN redundancy loss, security patching
- Or set up SNS notification with CloudWatch events
- Health > Specific Health Events
- github AWS Health AWS_RISK_CREDENTIALS_EXPOSED
- Can scan github looking for public IAM keys -> CloudWatch event -> Step functions that invoke lambdas to:
- Remove role from IAM
- Get audit from CloudTrail
- Output notifications to SNS
- KEY IDEA Use a STEP FUNCTION rather than just a Lambda, because it is much easier to debug problems or track progress