coredns won't start after updates ?

[olberger]

I've noticed tlat coredns doesn't seem to be starting with the latest minikube in tests I've made today:

NAMESPACE                NAME                                           READY   STATUS             RESTARTS   AGE
kube-system              pod/coredns-fb8b8dccf-7cp7q                    0/1     CrashLoopBackOff   22         68m
kube-system              pod/coredns-fb8b8dccf-pltdp                    0/1     CrashLoopBackOff   22         68m
...
NAMESPACE     NAME                                       READY   UP-TO-DATE   AVAILABLE   AGE
default       deployment.apps/antidote-web               1/1     1            1           19m
default       deployment.apps/nginx-ingress-controller   1/1     1            1           19m
default       deployment.apps/syringe                    1/1     1            1           19m
kube-system   deployment.apps/coredns                    0/2     2            0           68m

I'm using vagrant + libvirt...

Maybe the issue is other, but it happens that minikube just issued a new release...

Also, maybe the k8s version should be updated, as minikube seems to be in sync with 1.16 as per its release notes ?

Maybe the coredns service isn't used in Antidote, after all ?

Hope this helps.

[olberger]

Not sure which version of the box I was using, but I made another test with generic/ubuntu1804 and k8s 1.14.8 and this exhibits the problem...

Hth,

[olberger]

Latest Minikube seem to be working fine on generic/ubuntu1604 in this respect.

[Mierdin]

Is this still a problem or can we close this? I have yet to see this myself, especially recently.