[ricardobeat]

Only if you're crazy enough to put something in production running a single node instance.

[masklinn]

Even if you're not "crazy enough" to do what's prescribed, every user routed to the locked node instance will still be locked. You're just reducing the surface area of the freeze.

[ricardobeat]

Presumably not many requests will be routed to the locked instance - that's why it's called load "balancing".

And you shouldn't have any long-running computation on your server process anyway.

[rgbrgb]

I think this happens fairly often.