kubelet: new kubelet config option for disabling group oom kill#126096
Conversation
|
Skipping CI for Draft Pull Request. |
12311ee to
6e7d175
Compare
6e7d175 to
00f4125
Compare
|
For reviewers: I have just fixed the conflict in the previous PR because it had already been reviewed and was fine in itself. |
|
LGTM overall. Please address @roycaihw's comments above and fixed the failing tests. We are ready to go. /approve to unblock you. |
|
/test pull-kubernetes-node-kubelet-serial-containerd |
|
/test pull-kubernetes-e2e-capz-windows-master |
|
I guess |
Yes, I do. |
|
/test pull-kubernetes-node-kubelet-serial-containerd |
|
/lgtm Waiting for the test result: #126096 (comment) |
|
LGTM label has been added. DetailsGit tree hash: 02a3d5adf9dd34dc0ccf88bb694342898a14d5b9 |
|
/test pull-kubernetes-node-kubelet-serial-containerd I believe it will pass. |
liggitt
left a comment
There was a problem hiding this comment.
/approve
/hold for a nit on config field godoc to align with the yaml values users will specify, and one fix needed on the build tagging
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: dchen1107, liggitt, roycaihw, SergeyKanzhelev, utam0k, yujuhong The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
Signed-off-by: utam0k <k0ma@utam0k.jp>
|
/test pull-kubernetes-node-kubelet-serial-containerd |
|
@liggitt Thanks for your review. I've updated following your suggestions. |
|
/lgtm |
|
LGTM label has been added. DetailsGit tree hash: 1434b7ecb2606f35843efb840e693f15a8685a68 |
|
/test pull-kubernetes-unit |
What type of PR is this?
/kind feature
What this PR does / why we need it:
There is a difference in the OOM behavior between cgroup v1 and v2 by default. It's important to note that in cgroup v1 if one of the processes within a container is killed, the other processes will remain running. Ideally, we might want to align with cgroup v2's behavior, where all processes within a container are killed if an OOM event occurs. However, in reality, we need to maintain the behavior of cgroup v1 for the time being. Therefore, this pull request will introduce the
singleProcessOOMKillflag to enable like cgroup v1's behavior in cgroup v2.Appearently, this workaround has already been discussed in a SIG-Node meeting.
#117793 (comment)
Which issue(s) this PR fixes:
Carry: #122813 by @tzneal
Special notes for your reviewer:
Does this PR introduce a user-facing change?
Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.: