With 2 workers, they both get restarted instantly, almost every time.
With 3 workers, they sometimes get all restarted at once.
I used load balancing tool siege, with a 3 worker cluster, doing some "Graceful" restarts only resulted in 70% of requests being served. Native node-cluster module results in 100% availability so my script is fine.
With 2 workers, they both get restarted instantly, almost every time.
With 3 workers, they sometimes get all restarted at once.
I used load balancing tool
siege
, with a 3 worker cluster, doing some "Graceful" restarts only resulted in 70% of requests being served. Native node-cluster module results in 100% availability so my script is fine.