kubernetes/kops · critical

error loading configuration

Error message

error loading configuration %q: %v

What it means

nodeup's Run reads its boot configuration file from ConfigLocation via vfs.Context.ReadFile before doing anything else. This error wraps any failure to fetch that file: nonexistent path, unreadable object store, bad credentials, or network failure reaching the VFS location. nodeup cannot proceed without the boot config, so it aborts immediately.

Solutions

1. Verify the file exists at the ConfigLocation path (e.g. aws s3 ls s3://<bucket>/<path>)
2. Check the node's IAM instance profile grants read access to the state store location
3. Correct the --config flag / ConfigLocation value in the nodeup invocation or launch template
4. Confirm network connectivity from the node to the object store endpoint (VPC endpoints, proxy settings)

Example fix

// before
ConfigLocation: "s3://old-bucket/cluster.example.com/config/nodeupconfig"
// after
ConfigLocation: "s3://correct-state-bucket/cluster.example.com/config/nodeupconfig"

// before starting nodeup, confirm the config is readable
if err := exec.Command("aws", "s3", "cp", configLocation, "/dev/null").Run(); err != nil {
  return fmt.Errorf("config %s unreadable: %w", configLocation, err)
}

func hasConfigLocation(cmd *NodeUpCommand) bool { return cmd.ConfigLocation != "" }

if err := cmd.Run(os.Stdout); err != nil {
  if strings.HasPrefix(err.Error(), "error loading configuration") {
    klog.Errorf("cannot fetch boot config at %s; check path, IAM perms and network: %v", cmd.ConfigLocation, err)
    os.Exit(1)
  }
  return err
}

When it happens

Trigger: NodeUpCommand.Run with c.ConfigLocation set but the VFS path does not exist, points to a wrong bucket/cluster name, the instance lacks IAM permission to read it, or the S3/object-store endpoint is unreachable.

Common situations: Misconfigured nodeup systemd unit or ASG launch template passing a stale --config path; cluster state moved to a different S3 bucket; IAM instance profile missing s3:GetObject on the state bucket; DNS/proxy issues on a private network blocking S3.

Related errors

• cannot parse ConfigBase
• error loading NodeupConfig
• parsing path for kops-channels manifest
• Azure storage account ID is not set; it is required to…
• Azure storage account is not set on path

AI-assisted analysis of kubernetes/kops@4c8573c808 (2026-09-05). Data as JSON: /api/errors/227033641952e9d4. Report an issue: GitHub.

Appendix: source

Thrown at upup/pkg/fi/nodeup/command.go:92

// MaxTaskDuration is the amount of time to keep trying for; we retry for a long time - there is not really any great fallback
const MaxTaskDuration = 365 * 24 * time.Hour

// NodeUpCommand is the configuration for nodeup
type NodeUpCommand struct {
	CacheDir       string
	ConfigLocation string
	Target         string
}

// Run is responsible for perform the nodeup process
func (c *NodeUpCommand) Run(out io.Writer) error {
	ctx := context.Background()

	var bootConfig nodeup.BootConfig
	if c.ConfigLocation != "" {
		b, err := vfs.Context.ReadFile(c.ConfigLocation)
		if err != nil {
			return fmt.Errorf("error loading configuration %q: %v", c.ConfigLocation, err)
		}

		err = utils.YamlUnmarshal(b, &bootConfig)
		if err != nil {
			return fmt.Errorf("error parsing configuration %q: %v", c.ConfigLocation, err)
		}
	} else {
		return fmt.Errorf("ConfigLocation is required")
	}

	if c.CacheDir == "" {
		return fmt.Errorf("CacheDir is required")
	}

	region, err := getRegion(ctx, &bootConfig)
	if err != nil {
		return err
	}

View on GitHub (pinned to 4c8573c808)