daemon has %d active %s; setup saved the new configuration b

Error message

daemon has %d active %s; setup saved the new configuration but left the running daemon unchanged to avoid cancelling work. Wait for the active work to finish, then run '%s' to apply the new configuration

What it means

Deliberate safety refusal from the post-setup daemon restart logic: the daemon is alive, reports more than zero active tasks, and setup therefore refuses to restart it, because a restart would cancel in-flight work. The new configuration IS saved; only the running daemon still uses the old deployment. The message includes the exact restart command, profile-aware.

Source

Thrown at server/cmd/multica/cmd_setup.go:305

func dispatchDaemonAfterSetup(
	cmd *cobra.Command,
	args []string,
	health map[string]any,
	start setupDaemonRunner,
	restart setupDaemonRunner,
) error {
	if daemonAlive(health) {
		if activeTasks := daemonActiveTaskCount(health); activeTasks > 0 {
			taskLabel := "tasks"
			if activeTasks == 1 {
				taskLabel = "task"
			}
			restartCmd := "multica daemon restart"
			if profile := resolveProfile(cmd); profile != "" {
				restartCmd += " --profile " + profile
			}
			return fmt.Errorf(
				"daemon has %d active %s; setup saved the new configuration but left the running daemon unchanged to avoid cancelling work. Wait for the active work to finish, then run '%s' to apply the new configuration",
				activeTasks,
				taskLabel,
				restartCmd,
			)
		}
		fmt.Fprintln(os.Stderr, "\nRestarting daemon to apply the new configuration...")
		return restart(cmd, args)
	}

	fmt.Fprintln(os.Stderr, "\nStarting daemon...")
	return start(cmd, args)
}

func daemonActiveTaskCount(health map[string]any) int64 {
	switch count := health["active_task_count"].(type) {
	case float64:
		return int64(count)

View on GitHub (pinned to 2c0912b6ec)

Solutions

Wait for the reported active tasks to finish (check `multica daemon status`).
Run the exact command from the error message, e.g. `multica daemon restart --profile <P>`.
If the work must be abandoned, explicitly run `multica daemon restart` accepting the cancellation.
For unattended automation, schedule setup at idle times or drain tasks first.

Example fix

# before
multica setup self-host --server-url https://api.co --app-url https://app.co
# error: daemon has 3 active tasks ...
# after (once tasks drain)
multica daemon restart --profile default

Defensive patterns

Strategy: retry

Validate before calling

# Poll task count (single check, no busy loop) before restarting
multica daemon status --profile "$P" --output json | jq -e '.active_tasks == 0' >/dev/null || echo "tasks still active; not restarting"

Try / catch

Parse the count and restart command out of the message; treat as a soft-failure: pause the pipeline until tasks drain, then execute the suggested `multica daemon restart --profile <P>` and verify status.

Prevention

Don't force-restart daemons with active tasks unless cancellation is intended.
Automate setup at scheduled idle times.
Remember config IS saved at this point — only the restart is pending.

When it happens

Trigger: Running `multica setup` or `multica setup self-host` while the daemon has 1+ active tasks (per daemonActiveTaskCount from its health endpoint). Any deployment switch (cloud→self-host, server URL change) under active work triggers it.

Common situations: Re-pointing a busy workstation to a new server during work hours; CI agents with queued tasks; switching profiles while background sync/agents run.

Related errors

AI-assisted analysis of multica-ai/multica@2c0912b6ec (2026-08-15). Data as JSON: /api/errors/efc931017da431e7. Report an issue: GitHub.