Rollback from k8s 1.27.1 to 1.26.3 fails

Hi we have k8s 1.27.1 with coredns compatible version 1.10.0 and k8s 1.26.3 with compatible version of 1.9.3.
When we upgrade our cluster 1.26.3 => 1.27.1 it was fine. When we rollback from 1.27.1 => 1.26.3 we have below fatal error from kubeadm. Any pointers? Is this a k8s issue? It is referring somewhere to 1.10.0 though kubeadm config file is updated to use 1.9.3 during upgrade(rollback) plan/init/upgrade phase.

    "[upgrade/postupgrade] FATAL post-upgrade error: unable to get list of changes to the configuration.: start version '1.10.0' not supported",

Anyone has any inputs here please?

Hi Kiran,
It seems delete coredns deployment might be an workaround.
$kubectl delete deployment -n kube-system coredns
After that " /usr/local/bin/kubeadm upgrade apply" will be completed. the coredns deployment should be back after kubeadm upgrade .