AWS ENI Limits Constrain Max Pod Density on Smaller Instance Types
warningAWS CNI enforces low max pod counts (e.g., 58 for c6g.2xlarge) on instances with fewer CPUs due to ENI limitations, preventing full utilization of node compute capacity unless prefix delegation is enabled.
Monitor pod scheduling failures due to node pod capacity limits. Track instances where max pod count is significantly lower than CPU-based expectations. Watch for pod pending events with 'too many pods' reasons on nodes with <16 CPUs.
Enable AWS CNI prefix delegation with 'kubectl set env daemonset aws-node -n kube-system ENABLE_PREFIX_DELEGATION=true' to increase pod limits. Alternatively, set minimum CPU constraints in Cast AI Node Templates to 16+ CPUs to avoid affected instance types. Review AWS ENI limits per instance type at eni-max-pods.txt.