ignore that, i noticed it should woek

https://galaxy.ansible.com/holms/fqdn

on need task in linux.yml i will recommand to use :
notify: - restart hostname

and create a handlers : main.yml

• name: restart hostname
  service: name=hostname state=restarted
  async: 45
  poll: 0
  when: ansible_distribution not in [ 'RedHat', 'CentOS' ]

It's will avoid to always bounce the service and the final state result will be better ;)

What do you think ?

I get this message when using this:

[DEPRECATION WARNING]: "include" is deprecated, use include_tasks/import_tasks instead. See https://docs.ansible.com/ansible-core/2.14/user_guide/playbooks_reuse_includes.html for details. This feature will be removed in version 2.16.    
Deprecation warnings can be disabled by setting deprecation_warnings=False in ansible.cfg.

My playbook ran without errors but domainname still returns (none).

 ~/Code/home.b5y.gs  ✔  ssh home                                                                                                                  18.04 ● ↑3
Welcome to Ubuntu 18.04.1 LTS (GNU/Linux 4.15.0-29-generic x86_64)

 * Documentation:  https://help.ubuntu.com
 * Management:     https://landscape.canonical.com
 * Support:        https://ubuntu.com/advantage

  System information as of Fri Jan 25 15:02:06 GMT 2019

  System load:  0.1               Processes:              162
  Usage of /:   9.7% of 18.17GB   Users logged in:        1
  Memory usage: 1%                IP address for enp0s25: 10.0.1.254
  Swap usage:   0%

 * MicroK8s is Kubernetes in a snap. Made by devs for devs.
   One quick install on a workstation, VM, or appliance.

   - https://bit.ly/microk8s

 * Full K8s GPU support is now available!

   - https://blog.ubuntu.com/2018/12/10/using-gpgpus-with-kubernetes


 * Canonical Livepatch is available for installation.
   - Reduce system reboots and improve kernel security. Activate at:
     https://ubuntu.com/livepatch

162 packages can be updated.
88 updates are security updates.


Last login: Fri Jan 25 14:36:03 2019 from 10.0.1.30
brad@home:~$ hostname
home
brad@home:~$ domainname
(none)
brad@home:~$

Playbook:

---
# Set hostname and domain name locally
- hosts: all
  roles:
    - role: holms.fqdn
      fqdn: "{{fullname}}"

fullname is set to home.b5y.gs

any suggestions?

As per the title. tasks/linux.cf needs the following change:

lineinfile: dest=/etc/hosts line='{{ ansible_default_ipv4.address }} {{ fqdn }} {{ inventory_hostname_short }}' state=present backup=yes

should be

lineinfile: dest=/etc/hosts regexp='^{{ ansible_default_ipv4.address }}' line='{{ ansible_default_ipv4.address }} {{ fqdn }} {{ inventory_hostname_short }}' state=present backup=yes

If this is not done, a prexisting hosts entry (at first run) will take precedence over the line that ansible adds.

TASK [holms.fqdn : restart hostname] *******************************************
fatal: [xxx]: FAILED! => {"failed": true, "msg": "async mode is not supported with the service module"}

see ansible/ansible-modules-core#5472

maybe they will revert that change or fix it somehow.

Right now you specify both the FQDN and the hostname in the playbook. But for all of my cases, I just need this:

    - { role: holms.fqdn, fqdn: "{{ inventory_hostname }}", hostname: "{{ inventory_hostname_short }}" }

What about making that a default? Thoughts are appreciated.

- name: restart hostname
  service: name=hostname state=restarted
  async: 45
  poll: 0
  when: ansible_distribution not in [ 'RedHat', 'CentOS' ]

refer -

• ansible/ansible#18965
• ansible/ansible#18965 (comment)

Hi , to get correct value in /etc/hosts you should use {{ hostvars[item].ansible_hostname }}.{{ fqdn }}

it will print in /etc/hosts as

10.188.0.47 one.csc.fi one ( IP FQDN HOSTNAME )

use below in default.yml

  lineinfile: dest=/etc/hosts regexp='{{ hostvars[item].ansible_hostname }}$' line='{{ hostvars[item].ansible_default_ipv4.address }} {{ hostvars[item].ansible_hostname }}.{{ fqdn }} {{ item }}' state=present backup=yes

Hi,

The role failed on minimal debian if we connect without root account and via sudo ( if the dbus package is missing)

To fix it :

tasks/debian.yml

• name: Debian | Install dbus
  apt: name=dbus state=present

on the Debian Jessie the task restart hostname failed :

TASK [ansible-fqdn : restart hostname] *****************************************
fatal: [debian]: FAILED! => {"changed": false, "failed": true, "msg": "Unable to restart service hostname: Failed to restart hostname.service: Unit hostname.service is masked.\n"}

On the box, it's like this :

root@test:~# systemctl status hostname
● hostname.service
Loaded: masked (/dev/null)
Active: inactive (dead)

Having a .travis.yml file like this one makes testing a accepting pull request more easy. You can also specify versions that are know to fail (due to bugs).

Do you want me to add something similar?

Would you mind to start versioning the role and add tags? I would like to pin a specific version in my common setup role.

(absolutely necessary in a production environment)

Any interest in adding IPv6 support to this role? :)

The roles is failing on default Centos6 with this error :

fatal: [centos6]: FAILED! => {"changed": false, "failed": true, "msg": "Aborting, target uses selinux but python bindings (libselinux-python) aren't installed!"}

don't have this error on CentOS7 as they install by default the libselinux-python

i successfully tested this role on Ubuntu 16.04 LTS (xenial)

could you reimport it in the galaxy so xenial is in the list of supported versions?

I have a machine that has a separate Mgmt and Ops interface (and IP address).

I would need to set the IP of the fqdn entry with a custom address - not the ansible_default_ipv4.address.

This could solve #36 as one could set the address entry to an IPv6 address.

- hosts: 10.0.0.5
  roles:
    - { role: fqdn, fqdn: "mx.mydomain.com", hostname: "mx" }

/etc/hosts has the content:

10.0.0.5 mx.mydomain.com 10

but should have

10.0.0.5 mx.mydomain.com mx

I've only been using this role when deploying to a single host. Recently that changed, and after I fixed #3 I noticed this in my /etc/hosts:

127.0.0.1       localhost
127.0.1.1       ibm704
...
1.2.3.4 ibm704.example.com ibm704
4.3.2.1 ibm704.example.com therac25

Yikes 😲 My two hosts are ibm704.example.com and therac25.example.com. So the last line above should clearly be therac25.example.com instead of ibm704.example.com right? I just upgraded the role.