-
Notifications
You must be signed in to change notification settings - Fork 2.3k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
resctrl do not handle the orphan mon_groups #3057
Comments
another comment for resctrl comment |
Is that appropriate to deal with the issue in this way? Could you please help me to review itPR? Thanks. 🌹 @Creatone @katarzyna-z @dashpole |
Maybe the simplest solution is to clear all "cadvisor" prefixed mon groups during setup? |
It's the simplest solution. But when cadvisor crash frequently, it will lead to clear-up and re-create mon-group frequently. I'm not sure whether it will cost a lot for kernel as I'm not familar with the mon_group in the kernel 😭 . |
Frequently crashing is not expected behavior. |
@srl11 Could you check if that commit is sufficient for you? Creatone@3e2feae |
resctrl creates mon_group for container if cadvisor set resctrl_interval(!=0), and clear it when the container is destroyed.
But, cadvisor won't clear up the mon_group for the container which is destoryed during cadvisor restart.
So should we check the orphan mon_group, and clear it up?
releated version: v0.43.0 v0.42.0 v0.41.0 (commit)
My Solution:
if resctrl_interval!=0, cache the mon_group created by cadvisor, which has cadvisor monGroupPrefix, and clear up the expired mon_group.
@Creatone @katarzyna-z @dashpole
The text was updated successfully, but these errors were encountered: