Memory limit: requests/limits like in kubernetes

Incus
1

Is there a way to configure guaranteed limit and max limit for memory like requests/limits in kubernetes?

I think I need something working like:

  • hard limit, visible as memory limit inside CT via top/free and /proc/meminfo. OOM if hard limit is reached
  • guaranteed amount, no OOM if this limit is not reached, optional blocking creating new CT if “total guaranteed amount” of memory is overcommitted
2

That’s basically the limits.memory behavior, we set a soft limit of 90% of the amount and a hard limit at 100% of the amount, the OOM will trigger within the container if you use it all.

You can do that at the project level by setting a limits.memory on the project.
When you do that, every instance will need to have a limits.memory set so the total allocated (rather than used) can be computed, preventing you from creating instances once you’ve reached the project limit.

3

Do not understand idea about project limit.

The primary goal is not about preventing overcommitting, but about manageble guaranteed amount of memory level with no OOM.

4

The case is many CTs with sum(memory.limit) > node memory.

I’d like to set “no OOM amount” for some of them

5

Ah, right, so you want a two tier kinda limit for each instance, allowing to burst with extra risk.
I don’t believe there’s any way to do that at the kernel level on either cgroup1 or cgroup2.