Closed marquiz closed 2 months ago
Approved as it has nothing to do with this patch, either. I'm fine merging this patch and debugging/fixing the issue later.
There is no problem in latest Ubuntu 22.04.
It seems e2e tests fail on Ubuntu 24.04 due to container stop failure. From cri-resmgr.output.txt in vm:
E: [resource-manager] StopContainer: failed to stop container pod0:pod0c1: rpc error: code = Unknown desc = failed to stop container "97e4401acd43cba20543030364eb9986d579a886d5c9af5973cda5f2a3a24ae8": unknown error after kill: runc did not terminate successfully: exit status 1: unable to signal init: permission denied
I don't know if it has anything to do with cri-resmgr. Fails both topology-aware and balloons tests in the first
kubectl delete pod ...
command.
I'm guessing something SELinux-related, maybe... Probably failure to disable or set to permissive mode during node spinup ?
I'm guessing something SELinux-related, maybe
Does the latest ubuntu have SELinux enabled?
I'm guessing something SELinux-related, maybe
Does the latest ubuntu have SELinux enabled?
Argh... No, sorry, of course it does not. I was still in the context of the previous fedora-related PR I reviewed.
But this might be the cause: https://github.com/containerd/containerd/issues/10542
Rebased. PTAL @klihub
It seems e2e tests fail on Ubuntu 24.04 due to container stop failure. From cri-resmgr.output.txt in vm:
I don't know if it has anything to do with cri-resmgr. Fails both topology-aware and balloons tests in the first
kubectl delete pod ...
command.