Safe worker restart
Capability-aware restart, zero shell execution.
Celery broadcasts pool_restart, which is fire-and-forget and requires Celery's worker_pool_restarts setting; z4j does not receive confirmation that the pool restarted. Other supported adapters use self-exit and depend on an external supervisor to respawn the process, with documented in-flight-loss risk. Preflight requires both environment and filesystem signals, and a 60-second flap guard reduces restart loops.
Mockup of the dashboard view for this feature. Live-reloading and themed to match your dashboard.
Ships with
- Celery: pool_restart broadcast; observe the worker afterwards because delivery and restart are not confirmed
- Other supported adapters: self-exit plus supervisor respawn, with documented in-flight-loss risk
- Preflight = env signal + filesystem marker
- 60-second flap guard
- Never executes shell commands
Related