Interesting.
Perhaps you could try resolving service addresses from within a pod and
seeing if there's some other issue causing intermittent failures in
resolution.
The steps here
Abirudh,
Thanks for your response
I’m running k8s cluster on AWS and kub-dns pods are running fine and also
as I mentioned only 1 executor pod is running though I requested for 5 and
rest 4 were killed with below error and I do have enough resources
available.
On Tue, May 29, 2018 at 6:28 PM
This looks to me like a kube-dns error that's causing the driver DNS
address to not resolve.
It would be worth double checking that kube-dns is indeed running (in the
kube-system namespace).
Often, with environments like minikube, kube-dns may exit/crashloop due to
lack of resource.
On Tue, May