instanceGroup: %v does not exist (try adding --force flag)

Error message

instanceGroup: %v does not exist (try adding --force flag)

What it means

`kops replace` refuses to replace an InstanceGroup whose name is not found in the state store. Because `replace` is semantically an update operation, kOps conservatively fails instead of silently creating a brand-new InstanceGroup; passing `--force` explicitly opts in to create-on-missing behavior. The check is done via clientset.InstanceGroupsFor(cluster).Get, and the error is returned when errors.IsNotFound(err) is true and c.Force is false.

Source

Thrown at cmd/kops/replace.go:178

			case *kopsapi.InstanceGroup:
				clusterName := v.ObjectMeta.Labels[kopsapi.LabelClusterName]
				if clusterName == "" {
					return fmt.Errorf("must specify %q label with cluster name to replace instanceGroup", kopsapi.LabelClusterName)
				}
				cluster, err := clientset.GetCluster(ctx, clusterName)
				if err != nil {
					if errors.IsNotFound(err) {
						return fmt.Errorf("cluster %q not found", clusterName)
					}
					return fmt.Errorf("error fetching cluster %q: %v", clusterName, err)
				}
				// check if the instancegroup exists already
				igName := v.ObjectMeta.Name
				ig, err := clientset.InstanceGroupsFor(cluster).Get(ctx, igName, metav1.GetOptions{})
				if err != nil {
					if errors.IsNotFound(err) {
						if !c.Force {
							return fmt.Errorf("instanceGroup: %v does not exist (try adding --force flag)", igName)
						}
					} else {
						return fmt.Errorf("unable to check for instanceGroup: %v", err)
					}
				}
				switch ig {
				case nil:
					klog.Infof("instanceGroup: %v was not found, creating resource now", igName)
					_, err = clientset.InstanceGroupsFor(cluster).Create(ctx, v, metav1.CreateOptions{})
					if err != nil {
						return fmt.Errorf("error creating instanceGroup: %v", err)
					}
				default:
					_, err = clientset.InstanceGroupsFor(cluster).Update(ctx, v, metav1.UpdateOptions{})
					if err != nil {
						return fmt.Errorf("error replacing instanceGroup: %v", err)
					}
				}

View on GitHub (pinned to 4c8573c808)

Solutions

Re-run the command with the --force flag: `kops replace -f ig.yaml --force` to create the missing InstanceGroup.
Verify the `metadata.name` in the manifest matches an existing InstanceGroup (`kops get instancegroups <cluster>`) and fix the typo if not.
If the IG should already exist, confirm you are pointing at the right cluster/state store (--name and KOPS_STATE_STORE) before forcing creation.
Alternatively use `kops create -f ig.yaml` when the intent is creation, then `kops update cluster --yes` to apply.

Example fix

# before
kops replace -f instancegroup-nodes.yaml
// error: instanceGroup: nodes does not exist (try adding --force flag)
# after
kops replace -f instancegroup-nodes.yaml --force

Defensive patterns

Strategy: validation

Validate before calling

// Before replace, confirm the InstanceGroup exists
name := igManifest.ObjectMeta.Name
_, err := clientset.InstanceGroupsFor(cluster).Get(ctx, name, metav1.GetOptions{})
if errors.IsNotFound(err) {
    // either add --force or create first: kops create -f ig.yaml
    fmt.Printf("InstanceGroup %s does not exist; use --force or create it\n", name)
}

Prevention

Keep InstanceGroup manifest names in sync with `kops get instancegroups` output before replace
Use --force deliberately and only when creation is intended
Pin cluster name with --name to avoid checking the wrong state store
Script the existence check into CI before running replace

When it happens

Trigger: Running `kops replace -f ig.yaml` where the InstanceGroup name in the manifest (v.ObjectMeta.Name) does not exist in the cluster's state store and the --force flag was not supplied. The Get call returns a NotFound error, triggering this message.

Common situations: Typo in the InstanceGroup name in the manifest; renaming an existing InstanceGroup instead of editing it; replacing a manifest from a different/older cluster whose IGs were deleted; forgetting that replace does not create resources without --force.

Understand the failure class

Background: 'Could not be found', 'does not exist', 'not found in database': the resource-not-found family when an ID, slug, key, or URI lookup comes back empty — this error's family across 20 libraries.

Related errors

AI-assisted analysis of kubernetes/kops@4c8573c808 (2026-09-05). Data as JSON: /api/errors/7031dab61c52df5b. Report an issue: GitHub.