surya adapa:
Hey, Team at KodeKloud
was working with "Kubernetes the Hard way " setup on my local
must say, the walkthroughs are precisely what’s needed.
However, had encountered an issue, while deploying coredns and new pods on the cluster.
• it seems, the weave-cni needed plugins to be installed on the worker nodes to establish pod connectivity.
• have raised a git Pull Request : https://github.com/mmumshad/kubernetes-the-hard-way/pull/298
• this covers the Fix, and will help the new users have a seamless experience.
• also, this doesn’t require any update or mention in the <https://www.youtube.com/watch?v=HA0t7AsYzEA&list=PL2We04F3Y_41jYdadX55fdJplDvgNGENo&index=15|youtube videos>. The commands are pretty straight forward.
attaching snippets of the issue.
Alistair Mackay:
Hay @surya adapa
I will be sure to check this. Can you post the manifest of the pod you were trying to deploy?
I am shortly going to update it for 1.25, and CNI deps are in scope for upgrading as part of that.
unnivkn:
Hi @surya adapa could you please verify the weave-net pod logs & check it for any errors ?
surya adapa:
Hey @Alistair Mackay ^
as you mentioned, This is already covered in https://github.com/mmumshad/kubernetes-the-hard-way/blob/master/docs/09-install-cri-workers.md
It seems I actually missed executing these on worker nodes.
Thanks for Correcting.
surya adapa:
Hey @unnivkn,
https://github.com/mmumshad/kubernetes-the-hard-way/blob/master/docs/09-install-cri-workers.md
covers the same steps which fix it.
weave-net pods logs were all good, it was a case with every new pod(be it coredns or busybox). All were stuck in creation stage, with the above log lines[in screenshot]
