You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In combination with the cloud-controller-manager that handles the cloudy stuff, the networking should be a plain CNI bridge; set by a DaemonSet that drops the following file in /etc/cni/net.d on every node.
Since the kubenet->plain CNI stuff is proceeding, I want to test out how close we are using CNI-only (without the kubenet bells and whistles) on a cloud env like GCE.
I think the CNI spec will look something like this:
In combination with the cloud-controller-manager that handles the cloudy stuff, the networking should be a plain CNI bridge; set by a DaemonSet that drops the following file in
/etc/cni/net.d
on every node.Since the kubenet->plain CNI stuff is proceeding, I want to test out how close we are using CNI-only (without the kubenet bells and whistles) on a cloud env like GCE.
I think the CNI spec will look something like this:
(shamelessly inspired from @jbeda's project here: https://github.com/jbeda/kubeadm-gce-tf/blob/master/tf-scripts/prereq.sh#L46 😄)
The text was updated successfully, but these errors were encountered: