Timeout AWS GLOBAL Accelerator

When using AWS Global Accelerator with your ELB (Elastic Load Balancer), experiencing intermittent timeouts can be frustrating, especially if it works fine without the Global Accelerator. Here are some steps and considerations to troubleshoot and potentially resolve the issue:

Troubleshooting Steps

1. Check Health Checks:

• Ensure that the health checks for your endpoints (EC2 instances) configured in Global Accelerator are set correctly. If the health checks fail, Global Accelerator may not route traffic properly.
• Verify the health check settings in the Global Accelerator console and make sure they align with the health check settings of your ELB.

2. Review DNS Configuration:

• Make sure your DNS configuration for Global Accelerator is correct. Verify that the DNS records are pointing to the static IP addresses provided by Global Accelerator.
• Ensure that there are no conflicting DNS settings that might cause intermittent issues.

3. Check Network ACLs and Security Groups:

• Review the Network ACLs and Security Groups associated with your ELB and EC2 instances to ensure they allow traffic from Global Accelerator.
• Confirm that the required ports are open and that there are no rules that might be intermittently blocking traffic.

4. Monitor Logs and Metrics:

• Use CloudWatch to monitor the logs and metrics for both Global Accelerator and your ELB. Look for any anomalies or patterns that might indicate the cause of the timeouts.
• Specifically, check for metrics such as Rejected Connection Count, Active Flow Count, and Processed Bytes.

5. Verify Endpoint Group Configuration:

• Ensure that the endpoint groups in Global Accelerator are configured correctly and that they include all the necessary regions and endpoints.
• Check the traffic dial settings for each region to ensure that traffic is distributed as expected.

6. Test Latency and Connectivity:

• Use tools like curl or wget to test connectivity to your Global Accelerator IPs from different regions to identify if the issue is regional.
• Conduct latency tests to see if there are specific regions or endpoints that consistently have higher latency or timeouts.

Example Configuration Check

Here’s an example of how you might check your Global Accelerator configuration:

Health Check Configuration

Ensure the health checks for your Global Accelerator endpoints are set correctly:

{
    "HealthCheckProtocol": "HTTP",
    "HealthCheckPort": 80,
    "HealthCheckPath": "/health",
    "HealthCheckIntervalSeconds": 30,
    "ThresholdCount": 3
}

DNS Configuration

Verify your DNS settings to ensure they point to the correct static IPs provided by Global Accelerator.

Example CloudWatch Monitoring

Set up CloudWatch alarms to monitor critical metrics for Global Accelerator and ELB:

{
    "AlarmName": "HighRejectedConnectionCount",
    "MetricName": "RejectedConnectionCount",
    "Namespace": "AWS/GlobalAccelerator",
    "Statistic": "Sum",
    "Period": 60,
    "Threshold": 1,
    "ComparisonOperator": "GreaterThanThreshold",
    "EvaluationPeriods": 1,
    "AlarmActions": [
        "arn:aws:sns:us-east-1:123456789012:NotifyMe"
    ]
}

Additional Considerations

• Update Global Accelerator: Ensure you are using the latest configuration and consider re-creating the Global Accelerator if you suspect configuration corruption.
• Contact AWS Support: If the issue persists, consider reaching out to AWS Support for assistance. Provide them with detailed logs and metrics to help diagnose the problem.

By systematically checking these aspects, you should be able to identify and resolve the intermittent timeout issues with AWS Global Accelerator. If you need further assistance, providing detailed logs and configurations will be crucial for deeper analysis.