wait_for task fails about docker.ubuntu HOT 7 CLOSED

This is most likely related to how I am running the playbook. I setup the playbook as:

---
- name: Installing Docker
  hosts: all
  sudo: true
  roles:
    - angstwad.docker_ubuntu

Notice the sudo: true which is required for some of the tasks to run successfully. However this causes the wait_for task to fail on my host (OS X) because I'm not running ansible-playbook with sudo. The easiest fix would be to add sudo: false to the wait_for task like:

- name: Wait for instance to come online
  sudo: false
  local_action:
    module: wait_for
    host: "{{ ansible_ssh_host|default(inventory_hostname) }}"
    port: 22
    delay: 30
    timeout: 600
    state: started
  when: "(ansible_distribution_version == '12.04' and reboot_result|changed)
      or (ansible_distribution_version == '13.10' and cgroup_lite_result|changed)"

from docker.ubuntu.

So implementing this would be assuming that everyone is escalating privileges via sudo, where some might also be escalating privs via su.

I'll give this some thought, but I don't think implementing strictly sudo: false is the right way.

from docker.ubuntu.

Where are the docs for the su feature? I can't seem to find them.

from docker.ubuntu.

I don't think there are docs. But I wrote 98% of the su code in Ansible.

Yeah, if you're running the role as sudo and your remote sudo password
differs from your local sudo password, it'll fail.

There's probably a solution to this. I'm still investigating.

On Wednesday, June 11, 2014, Mark Aaron Shirley [email protected]
wrote:

Where are the docs for the su feature? I can't seem to find them.

—
Reply to this email directly or view it on GitHub
#9 (comment)
.

from docker.ubuntu.

For what it's worth, I hit this error because I run SSH on a non-standard port. I was able to fix it by adding to my ansible.cfg:

[defaults]
remote_port = 5678

Ideally docker.ubuntu would get this port out of inventory.ini, rather than just falling back to the default.

from docker.ubuntu.

@Wilfred Default behavior today is to pull the port from inventory, using ansible_ssh_port and defaulting to the defaults/main.yml#ssh_port value of 22. This may not have been pushed to Galaxy, so I'm updating the role there this morning.

- name: Wait for instance to come online
  local_action:
    module: wait_for
    host: "{{ ansible_ssh_host|default(inventory_hostname) }}"
    port: "{{ ansible_ssh_port|default(ssh_port) }}"
    delay: 30
    timeout: 600
    state: started
  when: "(ansible_distribution_version == '12.04' and reboot_result|changed)
      or (ansible_distribution_version == '13.10' and cgroup_lite_result|changed)"

from docker.ubuntu.

This issue has been open long enough -- I think the solution is to fork the role and add a sudo: false or su: false where necessary, since the options are mutually exclusive anyway and can't be added together on a single task.

from docker.ubuntu.