Comment # 4 on bug 1171770 from Francisco Freitas

(In reply to Thorsten Kukuk from comment #3)
> kured 1.3.0 is not the latest build, that contains kured 1.4.0.
> 
> I'm not sure where the problem is, you can try to use flannel instead of
> weave for the pod network. But flannel is not really maintained anymore and
> reports a lot of iptables errors.
> DNS isn't fully working, too.

Destroyed my cluster.
Done a transactional-update.
Same issue with the latest kured version : 
'''
time="2020-05-17T09:53:27Z" level=info msg="Kubernetes Reboot Daemon: 1.4.0"
time="2020-05-17T09:53:27Z" level=info msg="Node ID: worker01"
time="2020-05-17T09:53:27Z" level=info msg="Lock Annotation:
kube-system/kured:weave.works/kured-node-lock"
time="2020-05-17T09:53:27Z" level=info msg="Reboot Sentinel:
/var/run/reboot-required every 1h0m0s"
time="2020-05-17T09:53:27Z" level=info msg="Blocking Pod Selectors: []"
time="2020-05-17T09:53:27Z" level=info msg="Reboot on: SunMonTueWedThuFriSat
between 00:00 and 23:59 UTC"
'''

It does not come from kured. I got the same issue with multiple services (for
example haproxy-ingress).

For the CNI I tested :
- cilium (build a yaml for the github repository and put it in
/usr/shared/k8s-yaml/cilium)
- weavenet (default init)
- flannel (kubicctl init --pod-network flannel)


Same issue with all of them.