I'd forgotten how much I simply enjoy the opportunities for hacking - in the original naive sense of the word - that Kubernetes (K8s) offers.
Today I've been working to find out why CoreDNS didn't work in my cluster - clue, it was containerd that did it
However, I then started seeing: -
failed to allocate for range 0: no IP addresses available in range set: 10.48.131.1-10.48.131.62
from my CoreDNS pods, having "fixed"containerd.
Thankfully, Google Cloud have a doc for that: -
Pods display failed to allocate for range 0: no IP addresses available in range set error message
In part, it required me to stop containerd and kubelet, and clear out the previously defined IP address range: -
mkdir /var/lib/cni/networks
Once I did this, and restarted containerd and kubelet, we're back in the game !