Ubuntu, Ansible, and Docker walk into a traffic surge

[rgbkrk]

Deploying new nodes and I'm greeted with this friendly error:

TASK: [notebook | restart docker] *********************************************
failed: [tmp23-user.tmpnb.org] => {"failed": true}
msg:  * Docker is managed via upstart, try using service docker stop
 * Docker is managed via upstart, try using service docker start

failed: [tmp24-user.tmpnb.org] => {"failed": true}
msg:  * Docker is managed via upstart, try using service docker stop
 * Docker is managed via upstart, try using service docker start


FATAL: all hosts have already failed -- aborting

For now I ad-hoc'ed my way around it, but something has changed in either Ansible, Ubuntu, or Docker.

[ryanwalls]

+1, seeing the same thing with Docker 1.5.0 on Ubuntu 14.04

Using this task:

- name: restart docker
  service: name=docker state=restarted
  when: docker_service_state != 'stopped'

[alon-argus]

I believe this is related https://github.com/ansible/ansible-modules-core/issues/1170 (just bumped into this thread through google)