Health Checks

Monitoring and Failover Mechanisms

Endpoint Health Checks

Monitor the health of your application endpoints by sending HTTP/HTTPS/TCP requests.

HTTP/HTTPS/TCP protocols
Global health checkers
Configurable intervals

Calculated Health Checks

Combine results of multiple health checks using AND, OR, or NOT operations.

Boolean logic operations
Up to 256 child health checks
Complex health logic

CloudWatch Alarm

Monitor CloudWatch alarms and use their state to determine health status.

Based on CloudWatch metrics
Custom application metrics
Private resource monitoring

Health Check Configuration

Request Interval

30 seconds (standard) or 10 seconds (fast)

  • • Standard: Lower cost
  • • Fast: Quicker failover
  • • Global health checkers

Failure Threshold

Number of consecutive failures before unhealthy

  • • Default: 3 failures
  • • Range: 1-10 failures
  • • Configurable threshold

String Matching

Search for specific string in response

  • • First 5120 bytes
  • • Case-sensitive search
  • • Optional feature

Notifications and Actions

SNS Notifications

Health check state changes
Email, SMS, HTTP endpoints
Real-time alerts

Automatic Failover

DNS record updates
Traffic rerouting
High availability

Health Checks Exam Tips

  • • Health checks are required for failover routing policy
  • • Endpoint health checks monitor HTTP/HTTPS/TCP endpoints
  • • Calculated health checks combine multiple health checks with boolean logic
  • • CloudWatch alarm health checks monitor private resources
  • • Health checkers are distributed globally for redundancy
Previous: Routing Policies Next: Domain Registration