kubernetes/kops · error

errShutdown

errShutdown

Error message

the client is shutdown

What it means

Returned by findAutoscalingGroupLaunchTemplate (used by rolling-update and instance-group operations) when an ASG exposes neither a direct LaunchTemplate nor a MixedInstancesPolicy launch template specification. kOps requires one of these to compute the launch template ID; launch configurations (legacy) are not accepted by this path.

Source

Thrown at pkg/otel/otlptracefile/client.go:87

}

// Stop implements otlptrace.Client.
func (c *client) Stop(ctx context.Context) error {
	c.writerMutex.Lock()
	defer c.writerMutex.Unlock()

	if c.writer != nil {
		err := c.writer.Close()
		if err != nil {
			return err
		}
		c.writer = nil
	}

	return nil
}

var errShutdown = errors.New("the client is shutdown")

// UploadTraces implements otlptrace.Client.
func (c *client) UploadTraces(ctx context.Context, protoSpans []*tracepb.ResourceSpans) error {
	c.writerMutex.RLock()
	defer c.writerMutex.RUnlock()

	if c.writer == nil {
		return errShutdown
	}

	return c.writer.writeTraces(ctx, &coltracepb.ExportTraceServiceRequest{
		ResourceSpans: protoSpans,
	})
}

// MarshalLog is the marshaling function used by the logging system to represent this Client.
func (c *client) MarshalLog() interface{} {
	return struct {

View on GitHub (pinned to 4c8573c808)

Solutions

1. Migrate the cluster to launch templates: `kops update cluster` with Kubelet/IG spec on a kOps version defaulting to launch templates
2. Replace the ASG via `kops rolling-update cluster --cloudonly` after switching to launch templates
3. Check for externally-managed ASGs tagged with the cluster tags and remove them
4. If you must support launch configurations, patch the fallback path to use DescribeLaunchConfigurations

Example fix

// before
} else {
	return "", fmt.Errorf("error finding launch template or configuration for autoscaling group: %s", aws.ToString(g.AutoScalingGroupName))
}
// after
} else if g.LaunchConfigurationName != nil {
	// legacy launch-configuration ASGs are not supported for LT lookup
	return "", fmt.Errorf("ASG %s uses a launch configuration; migrate to launch templates", aws.ToString(g.AutoScalingGroupName))
} else {
	return "", fmt.Errorf("error finding launch template or configuration for autoscaling group: %s", aws.ToString(g.AutoScalingGroupName))
}

if g.LaunchTemplate == nil && g.MixedInstancesPolicy == nil {
  return fmt.Errorf("ASG %q has no launch template; migrate from launch configurations", aws.ToString(g.AutoScalingGroupName))
}

func usesLaunchTemplate(g *autoscalingtypes.AutoScalingGroup) bool {
  return g.LaunchTemplate != nil || (g.MixedInstancesPolicy != nil && g.MixedInstancesPolicy.LaunchTemplate != nil && g.MixedInstancesPolicy.LaunchTemplate.LaunchTemplateSpecification != nil)
}

err := rollingUpdate(ctx, cluster)
if err != nil && strings.Contains(err.Error(), "error finding launch template or configuration") {
  // migrate the ASG to launch templates via kops update, then retry
  migrateToLaunchTemplates(); retry()
}

When it happens

Trigger: ASG was created with a classic LaunchConfiguration, or the ASG description returned nil for both LaunchTemplate and MixedInstancesPolicy (e.g. externally created ASG or an unexpected AWS API shape).

Common situations: Cluster provisioned by an older kOps version using launch configurations; ASG hand-edited or created by Terraform with LaunchConfigurationName; partial MixedInstancesPolicy responses.

Related errors

• DIGITALOCEAN_ACCESS_TOKEN is required
• expected a aws.Cloud provider
• timed out waiting for volume to detach
• failed to detach instance: %v
• could not find one of launch template or mixed instances pol

AI-assisted analysis of kubernetes/kops@4c8573c808 (2026-09-05). Data as JSON: /api/errors/18c2246ef5c289f0. Report an issue: GitHub.