# First, let's examine the actual file and its context
cat -n prod/eks/terragrunt.hcl | head -40

# Search for any exception approvals or justification comments in the repo
rg -i "exception|approval|t3.*approved|graviton.*first|instance.*policy" -A 3 -B 3

# Look for container image build configuration
fd -e Dockerfile -e "*.yml" -e "*.yaml" | head -20

# Search for any documentation about instance type policy or requirements
rg -i "instance.*type.*policy|graviton.*first|aws.*standard|t3.*prohibited|t4g.*required" -g '*.md' -g '*.txt' -g '*.rst'

# Check git history for justification of this change
git log --oneline -20 --all -- prod/eks/terragrunt.hcl 2>/dev/null || echo "Git history not available"

# Search for multi-arch or platform-specific image build configuration
rg -i "docker buildx|platform.*linux|arm64|amd64|multi.*arch" -g 'Dockerfile*' -g '*.yml' -g '*.yaml' -g '*.sh' | head -30

# Check git commit details for the t3 migration
git show bfe9a34 --stat 2>/dev/null || echo "Cannot show full commit"

# Search for any README or documentation about instance type requirements
fd -e README.md -e '*.md' -e '*.txt' | xargs grep -l "instance\|t3\|t4g\|compatibility" 2>/dev/null || echo "No docs found"

# Check if there are any other node group configurations for comparison
rg "instance_types|arm_instance" -B 3 -A 3

# Look for any comments or variables that might explain the compatibility issue
rg "compatibility|t3|t4g" --context=5

In prod/eks/terragrunt.hcl around lines 25-26, the variable arm_instance_types
was set to x86 t3 instances which conflicts with the EKS module configured as
ami_type = "AL2_ARM_64"; fix by either (A) reverting to ARM instance types
(e.g., "t4g.large","t4g.xlarge") to match AL2_ARM_64 and keep variable name, or
(B) change the AMI to AL2_x86_64 and rename the variable from arm_instance_types
to instance_types (and rename related variables like arm_desired_size →
desired_size) to reflect x86 usage; also update README and any module/variable
docs to match the chosen architecture and amend the commit message to explain
which workloads required the change and why this instance/AMI choice resolves
that need.