Getting "Kubenet does not have netConfig. This is most likely due to lack of PodCIDR" as of today

[Zimmergren]

Hi,

Another issue that keeps appearing since yesterday/this morning is:

network is not ready: [runtime network not ready: NetworkReady=false reason:NetworkPluginNotReady message:docker: network plugin is not ready: Kubenet does not have netConfig. This is most likely due to lack of PodCIDR]

This happens on one of my Daemon Sets; After deleting the daemon set and re-deploying it, I got the same issue again.

My other deployments work - and this has worked all up until yesterday.

I have not:

• Done any upgrades or maintenance work on the cluster
• Changed any configurations in the daemonset, deployments or other resources in AKS and Kube
• Done anything else other than monitoring my cluster using Log Analytics

I have:

• Scaled the cluster from N nodes to N+1 nodes
• The new node has a state of "Ready: Unknown", likely causing the above.

TLDR:

• Do we have more known issues with node scaling?
• Every second/third time I scale my cluster, it breaks with random issues like this one.
• Estimated about every ~10th time I do scaling, I have to delete the entire cluster and start over due to the handshake timeouts, kubectl and cluster unresponsiveness and general inaccessibility of the cluster after triggering a node scale.

[Zimmergren]

I've reproduced and verified that this issue happens when this other issue (#274) is happening. Closing this.