Ole Markus With
Ole Markus With
I see the exact same error running `kubectl exec` or `ctr task exec` against a container when Kubelet is using `cpuManagerPolicy: static`
I can confirm runc 1.1.2 does not have this problem. Thanks for the tip.
Does this belong with runc project? Is there an issue filed there already?
> One thing I don't understand is how exactly `systemctl daemon-reload` affects things. Red herring, I'm thinking. You don't have to run that to reproduce.
> Maybe it's obvious, but I haven't seen it mentioned: If I leave out the `-t` option for the `exec` command, it works without error. Of course using a shell...
FWIW, EKS-D + kOps can easily be deployed on openstack on-premise using a guide similar to the AWS guide.
@dprateek1991 it does. It is just "beta" support as we currently do not run e2e tests, but it is actively maintained and there are many using it in prod. See...
As mentioned, this is needed for IRSA. See https://kubernetes.io/docs/tasks/configure-pod-container/configure-service-account/#service-account-token-volume-projection
Being able to have LB controller manage either instance or IP targets to a global accelerator would be really useful.