http://bugzilla.opensuse.org/show_bug.cgi?id=1198064
http://bugzilla.opensuse.org/show_bug.cgi?id=1198064#c4
--- Comment #4 from Johannes Kastl ---
Hmmm, I think I have to eat my words.
The cluster using flannel appears to be working, i.e. all pods are starting up.
But I still see lots of errors in the coredns pod and DNS resolution to outside
of the cluster is still broken.
I.e. my registry pod cannot connect to registry-1.docker.io, my rancher
cluster-agent pod cannot connect to the Rancher instance running in the local
network, as it fails to resolve the DNS name.
The other cluster that I tried to fix by switching to flannel also does not
work properly:
[ERROR] plugin/errors: 2 git.rancher.io. AAAA: read udp
10.61.0.4:46457->192.168.99.202:53: i/o timeout
[ERROR] plugin/errors: 2 git.rancher.io. A: read udp
10.61.0.4:44259->192.168.99.202:53: i/o timeout
[ERROR] plugin/errors: 2 registry-1.docker.io. A: read udp
10.61.0.4:33801->192.168.99.152:53: i/o timeout
[ERROR] plugin/errors: 2 registry-1.docker.io. AAAA: read udp
10.61.0.4:55431->192.168.99.202:53: i/o timeout
Not sure if I broke something else when switching to flannel?
Enno, can you check your coredns pod logs for errors, please? I would like to
know why it apparently works in your cluster but mine are broken...
Kind Regards,
Johannes
--
You are receiving this mail because:
You are the assignee for the bug.
