Add CloudWatch Alarm for Cerberus execution failure

[gregsienkiewicz]

Add monitoring and alerting capabilities to Cerberus

Summary

Enhanced the Cerberus state machine with comprehensive monitoring and alerting functionality to ensure quick response to execution failures.

Changes Made

New CloudFormation Resources

• CloudWatch Alarm (CerberusExecutionFailureAlarm): Monitors state machine for execution failures
• SNS Topic (CerberusFailureNotificationTopic): Handles failure notifications via email

New Template Parameter

• NotificationEmail: Required parameter for specifying email address to receive failure alerts
  • Includes validation pattern for proper email format
  • Used for SNS topic subscription

Key Features

• Automatic Failure Detection: CloudWatch alarm triggers on any state machine execution failure
• Immediate Notifications: SNS sends email alerts to specified address when failures occur
• Zero Configuration: Monitoring is automatically enabled upon deployment

Breaking Changes

⚠️ BREAKING: The NotificationEmail parameter is now required for deployment. Existing deployments will need to include this parameter in their next update.

Migration Guide

When updating existing deployments, add the NotificationEmail parameter:

sam deploy --parameter-overrides NotificationEmail=your-email@company.com [other-params]