kubernetes/kops · error

unknown load balancer Type

Error message

unknown load balancer Type: %q

What it means

During AWS model building, kOps maps the cluster spec's api.loadBalancer.type to an ELBv2 scheme. Only the enum values 'Internal' and 'Public' are accepted; anything else makes BuildNodePortTask/Build fail with this error rather than guessing a scheme.

Solutions

1. Set spec.api.loadBalancer.type in the cluster spec to exactly "public" or "internal" (kops accepts only these kops.LoadBalancerType values, lowercase).
2. If the field is empty in cluster.yaml, run `kops edit cluster` and add the type explicitly, e.g. `type: Public`.
3. Check for casing/typo with `kops get cluster -oyaml | grep -A3 loadBalancer` and fix to lowercase internal/public.
4. Upgrade kOps if a previously valid value stopped working; older/newer schemas may differ — validate with `kops toolbox template` or `kops replace -f` to surface schema errors early.

Example fix

// before (cluster.yaml)
api:
  loadBalancer:
    type: External
// after
api:
  loadBalancer:
    type: Public

# validate cluster spec before kops update
type=$(kops get cluster -oyaml | awk '/^[ ]*type:/{print $2}')
case "$type" in
  Public|public|Internal|internal) echo "ok";;
  *) echo "invalid loadBalancer type: '$type' (must be public or internal)"; exit 1;;
esac

When it happens

Trigger: `kops update cluster` (or `kops create cluster --api-loadbalancer-type=...`) with spec.api.loadBalancer.type set to a value other than "internal" or "public" (empty string, typo like "internall"/"external", or wrong casing, since the switch is case-sensitive).

Common situations: Hand-edited cluster.yaml with a misspelled or capitalized type; copying config from a classic-ELB-era docs page; tooling that serializes the field as "" when unset; migrating configs where the field was removed/renamed.

Understand the failure class

Background: Invalid enum value errors: "Unknown type", "Invalid scope", "must be one of" — when a string is not on the library's allowed list — this error's family across 23 libraries.

Related errors

• spotinst: load balancer
• AttachDetachReconcileSyncPeriod cannot be set to less than…
• AWS partition was empty
• Azure storage account name is required
• building presigned request

AI-assisted analysis of kubernetes/kops@4c8573c808 (2026-09-05). Data as JSON: /api/errors/ee4778404dcb50fa. Report an issue: GitHub.

Appendix: source

Thrown at pkg/model/awsmodel/api_loadbalancer.go:228

		if b.Cluster.UsesLoadBalancerForKopsController() {
			nlb.SetWaitForLoadBalancerReady(true)
		}

		if b.Cluster.UsesLoadBalancerForKopsController() {
			lbSpec.CrossZoneLoadBalancing = new(true)
		} else if lbSpec.CrossZoneLoadBalancing == nil {
			lbSpec.CrossZoneLoadBalancing = new(false)
		}

		nlb.CrossZoneLoadBalancing = lbSpec.CrossZoneLoadBalancing

		switch lbSpec.Type {
		case kops.LoadBalancerTypeInternal:
			nlb.Scheme = elbv2types.LoadBalancerSchemeEnumInternal
		case kops.LoadBalancerTypePublic:
			nlb.Scheme = elbv2types.LoadBalancerSchemeEnumInternetFacing
		default:
			return fmt.Errorf("unknown load balancer Type: %q", lbSpec.Type)
		}

		if lbSpec.AccessLog != nil {
			nlb.AccessLog = &awstasks.NetworkLoadBalancerAccessLog{
				Enabled:        new(true),
				S3BucketName:   lbSpec.AccessLog.Bucket,
				S3BucketPrefix: lbSpec.AccessLog.BucketPrefix,
			}
		} else {
			nlb.AccessLog = &awstasks.NetworkLoadBalancerAccessLog{
				Enabled: new(false),
			}
		}

		{
			groupAttrs := map[string]string{
				awstasks.TargetGroupAttributeDeregistrationDelayConnectionTerminationEnabled: "true",
				awstasks.TargetGroupAttributeDeregistrationDelayTimeoutSeconds:               "30",

View on GitHub (pinned to 4c8573c808)