r6i.xlarge instances are intermittently failing to launch in eu-west-2

Based on the system logs you've provided, this appears to be a networking initialization issue where the eth0 interface is not being found during instance boot. While intermittent launch failures can occur due to insufficient instance capacity in specific Availability Zones, the specific error you're seeing suggests a different root cause.

For intermittent launch issues in a specific region, here are some troubleshooting steps:

1. Check the State Transition Reason: In the EC2 console, select the failed instance and look at the "State transition reason" field on the first tab. This will provide specific details about why the instance failed.

2. Try Different Availability Zones: Since you mentioned this is specific to eu-west-2, try launching instances across different Availability Zones within that region. Sometimes capacity or configuration issues can be isolated to specific zones.

3. Verify Instance Configuration: Ensure your launch template or configuration doesn't have any region-specific settings that might be causing conflicts with network initialization.

4. Consider Alternative Instance Types: If this is specific to r6i.xlarge instances, temporarily try launching a different instance type in the same family (like r6i.large or r6i.2xlarge) to see if the issue persists.

5. Review AMI Compatibility: Verify that your AMI is fully compatible with the r6i instance type and the eu-west-2 region, particularly regarding network driver support.

Given the specific nature of the networking error and its regional isolation, I'd recommend opening a support case with AWS to investigate whether there's an underlying infrastructure issue in eu-west-2 affecting r6i instances specifically.
Sources
Troubleshoot Amazon EC2 instance launch issues - Amazon Elastic Compute Cloud
Troubleshoot Amazon EC2 Auto Scaling: EC2 instance launch failures - Amazon EC2 Auto Scaling