Hacker News Reader

Does it not strike anyone else as odd that if someone said they had a single CPU, and that CPU were running a normal priority task at 100%, and that caused the machine to barely allow ssh, we'd say there's a much bigger problem than that someone is letting on?

No 32 core (thread, likely) machine should ever normally be in a state where someone can "barely ssh onto it". Is Java really that janky? Or is "barely ssh onto it" a bit hyperbolic?

View on HN

I've absolutely experienced situations like this where the system is maxed out and ssh becomes barely usable.

The problem is that all processes run with pretty much the same priority level, so there's no differentiation in the scheduler between interactive ssh sessions and other programs that are consuming all the CPU, and most of the time you only realize this fact far after the point where you can SSH in and renice the misbehaving processes.

My macOS install got into a state before the Christmas holiday where something would spawn #corecount amount of process. They seemed to be related to processing thumbnails for contents of directories.

This would go on for a few minutes in which the machine was completely unresponsive including the mouse.

So it seems possible still to bring a high core count machine to its knees. But something then is indeed very wrong.

Reply on news.ycombinator.com