kubernetes/kops · error

error creating cluster: %v

Error message

error creating cluster: %v

What it means

Wraps a failure of clientset.CreateCluster while `kops replace --force` was creating a brand-new cluster in the state store. The create call (not the config parsing) failed; the underlying API/registry error is carried in %v.

Source

Thrown at cmd/kops/replace.go:150

						if errors.IsNotFound(err) {
							cluster = nil
						} else {
							return fmt.Errorf("error fetching cluster %q: %v", clusterName, err)
						}
					}
					if cluster == nil {
						if !c.Force {
							return fmt.Errorf("cluster %v does not exist (try adding --force flag)", clusterName)
						}

						err = cloudup.PerformAssignments(v, vfsContext, cloud)
						if err != nil {
							return fmt.Errorf("error populating configuration: %w", err)
						}

						_, err = clientset.CreateCluster(ctx, v)
						if err != nil {
							return fmt.Errorf("error creating cluster: %v", err)
						}
					} else {
						_, err = clientset.UpdateCluster(ctx, v, status)
						if err != nil {
							return fmt.Errorf("error replacing cluster: %v", err)
						}
					}
				}

			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)

View on GitHub (pinned to 4c8573c808)

Solutions

1. Check write permissions on the state store bucket/container and credentials
2. Re-run with `kops get clusters` to see if the cluster now exists (race) and use replace/update instead
3. Validate the manifest spec (`kops replace` decodes strictly; fix invalid fields like kubernetesVersion, networking)
4. Confirm the state store is healthy (no lifecycle rules deleting objects, no SSE/KMS access issues)
5. Read the wrapped %v error for the exact backend or validation cause

Example fix

// before
aws s3api put-object --bucket my-state-store --key test  # AccessDenied
// after
# fix IAM policy to allow s3:PutObject on the state store bucket
kops replace -f cluster.yaml --force

// ensure state store is writable
key := path.Join(stateStore, "test-write-probe")
if err := writeProbe(stateStore); err != nil {
    return fmt.Errorf("state store not writable: %w", err)
}

_, err := clientset.CreateCluster(ctx, v)
if err != nil {
    if isConflict(err) {
        // cluster created concurrently: fall back to update path
    } else if isTransient(err) {
        // retry with backoff
    }
    return err
}

When it happens

Trigger: CreateCluster failing due to an unwritable state store (permissions, quota), an invalid cluster spec rejected by validation, a conflict because the object was concurrently created, or serialization issues with the API version.

Common situations: State store bucket is read-only or credentials lack write access; bucket region/permissions changed; cluster was created by another process between the Get and Create; malformed spec (e.g. invalid kubernetesVersion or networking config).

Related errors

• error storing InstanceGroup: %v
• error replacing cluster: %v
• cluster not found %q
• reading instance groups: %w
• reading file %v: %w

AI-assisted analysis of kubernetes/kops@4c8573c808 (2026-09-05). Data as JSON: /api/errors/b6eab23b100a79ee. Report an issue: GitHub.