ssh connection is broken when creating kubernates cluster on environment which has vlan using 172.17.0.1 /16 cidr.

[Gökhan]

Sometimes we can use 172.17.0.1/16 cidr on a vlan in our envrionment. When try to create kubernates cluster on this environment, ssh connection to kubemaster is broken because when boot images it creates docker automaticaly with docker0 bridge with 172.17.0.1/16 cidr. Docker is using flannel network so in fact it shouldn't lost ssh connection. then I see that it is order bug. services are started before network config in https://github.com/openstack/magnum/blob/master/magnum/drivers/k8s_fedora_atomic_v1/templates/kubemaster.yaml#L490
so my solution is to write - config: {get_resource: enable_services} line after the https://github.com/openstack/magnum/blob/master/magnum/drivers/k8s_fedora_atomic_v1/templates/kubemaster.yaml#L496 this line.

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/pike
Review: https://review.openstack.org/524151

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: stable/ocata
Review: https://review.openstack.org/524152

[Spyros Trigazis (strigazi)]

When you boot a fedora atomic image without magnum. Do you see the same problem?

[OpenStack Infra (hudson-openstack)]

Change abandoned by Spyros Trigazis (<email address hidden>) on branch: stable/pike
Review: https://review.opendev.org/524151
Reason: The magnum team is cleaning up the backlog of changes older than 30 days. Feel to restore your patch.

[OpenStack Infra (hudson-openstack)]

Change abandoned by Spyros Trigazis (<email address hidden>) on branch: master
Review: https://review.opendev.org/523837
Reason: The magnum team is cleaning up the backlog of changes older than 30 days. Feel to restore your patch.

[OpenStack Infra (hudson-openstack)]

Change abandoned by Spyros Trigazis (<email address hidden>) on branch: stable/ocata
Review: https://review.opendev.org/524152
Reason: The magnum team is cleaning up the backlog of changes older than 30 days. Feel to restore your patch.