GoogleContainerTools/skaffold · warning

cleaning up skaffold created network

Error message

cleaning up skaffold created network

What it means

If the deployer created a dedicated Docker network, Cleanup removes it with d.client.NetworkRemove, but only when d.networkDeployed is true. A failure removing the network (endpoints still attached, already gone, daemon error) is wrapped with this message.

Source

Thrown at pkg/skaffold/deploy/docker/deploy.go:378

		}
		return nil
	}
	for _, container := range d.tracker.DeployedContainers() {
		if err := d.client.Delete(ctx, out, container.ID); err != nil {
			// TODO(nkubala): replace with actionable error
			return errors.Wrap(err, "cleaning up deployed container")
		}
		d.portManager.RelinquishPorts(container.ID)
	}

	for _, m := range d.debugger.SupportMounts() {
		if err := d.client.VolumeRemove(ctx, m.Source); err != nil {
			return errors.Wrap(err, "cleaning up debug support volume")
		}
	}

	if err := d.client.NetworkRemove(ctx, d.network); d.networkDeployed && err != nil {
		return errors.Wrap(err, "cleaning up skaffold created network")
	}

	return d.cleanPreviousDeployments(ctx)
}

func (d *Deployer) cleanPreviousDeployments(ctx context.Context) error {
	ctd, err := d.containersToDelete(ctx)
	if err != nil {
		return err
	}

	ntd, err := d.networksToDelete(ctx, ctd)
	if err != nil {
		return err
	}

	vtd := d.volumesToDelete(ctd)

View on GitHub (pinned to a1189de023)

Solutions

1. Disconnect/remove remaining containers first: `docker network inspect <network>` to see attached endpoints
2. Run `docker network rm <network>` manually, or `docker network prune`
3. Re-run `skaffold delete` once the container cleanup succeeds
4. If the network is already gone, the leftover error is harmless — ignore or upgrade skaffold to tolerate 404
5. Avoid running concurrent skaffold sessions against the same network

Example fix

// before
error: cleaning up skaffold created network: Error response from daemon: network <id> has active endpoints
// after
$ docker network inspect <network>   # find endpoints
$ docker rm -f <endpoint-container>
$ docker network rm <network>
$ skaffold delete

if out, err := exec.Command("docker", "network", "inspect", d.network).CombinedOutput(); err != nil {
  log.Printf("network already absent or uninspectable: %s", out) // nothing to clean
}

null

if err := deployer.Cleanup(ctx, out); err != nil && strings.Contains(err.Error(), "cleaning up skaffold created network") {
  log.Printf("network cleanup deferred: %v", err)
  exec.Command("docker", "network", "prune", "-f").Run()
}

When it happens

Trigger: Deployer.Cleanup reaches d.client.NetworkRemove(ctx, d.network) with d.networkDeployed == true and the daemon returns an error — typically because containers are still connected to the network or it was already deleted.

Common situations: Containers still attached because earlier cleanup steps (containers) failed; network manually deleted with `docker network rm`; two skaffold instances sharing one network race to delete it; daemon down during cleanup.

Related errors

• cleaning up skaffold created network
• INIT_DOCKER_NETWORK_CONTAINER_DOES_NOT_EXIST
• creating skaffold network %s: %w
• pruning removed container: %w
• unable to stream build output: %w

AI-assisted analysis of GoogleContainerTools/skaffold@a1189de023 (2026-09-05). Data as JSON: /api/errors/be72102d0ec8701e. Report an issue: GitHub.