Skip to content

Conversation

@askervin
Copy link
Collaborator

A node with exactly enough capacity should win a node without enough capacity.

A node with exactly enough capacity should win a node without
enough capacity.

Signed-off-by: Antti Kervinen <antti.kervinen@intel.com>
@klihub
Copy link
Collaborator

klihub commented Nov 26, 2025

I think this should only be allowed if there are not containers assigned to the shared subset of CPUs in the pool. Otherwise we might use up the shared cpuset fully and try to reassign burstable containers in the pool to the empty cpuset which will then fail.

@klihub
Copy link
Collaborator

klihub commented Dec 4, 2025

This is now fixed by PR #601. Closing this in favor of that.

@klihub klihub closed this Dec 4, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants