Ansible - unable to append correctly using the file plugin - ansible

I'm trying to use the Ansible file plugin to append data to a txt file.
This is my playbook:
- name: Setting some variables.
hosts: "{{ target }}"
gather_facts: no
connection: local
tasks:
- name: Delete newDevices file.
file:
path: newDevices.txt
state: absent
run_once: true
- name: Create newDevices file.
file:
path: newDevices.txt
state: touch
run_once: true
- name: Add new devices to file.
delegate_to: 127.0.0.1
lineinfile:
insertafter: EOF
path: newDevices.txt
line: "{{ inventory_hostname }}"
This is my host file:
[lab]
routerA01.mgt.net
routerA02.mgt.net
routerB01.mgt.net
routerB02.mgt.net
This is how I'm running the playbook:
ansible-playbook myplaybook.yml -i hosts --limit "lab[0-1]" -e target=lab
The newDevices.txt file gets deleted and recreated but the contents of the file vary from having 1 hostname or 2 hostnames. If I change the limit in the run command to --limit "lab[0-2]" then the same thing happens. Sometimes 2 hostnames will print and other times all 3 will print. Seems like there will always be a time when the newDevices.txt file will contain one less hostname in it.
Not sure why this is happening. Tried adding a pause after the file is created to maybe give the processing a little extra time, but that didn't help either.

There is a couple of improvements
Put all tasks into a block delegated to localhost. All of them should run only once on the localhost anyway.
Then, there is no point to declare global connection: local
Set hosts: all if you plan to limit the inventory on the command line. In this case, there is no point to put the group into a variable.
Write the file in a loop. This avoids concurrent writing from multiple instances. (What is probably the main reason for your troubles.)
Given the (simplified) inventory
shell> cat hosts
[lab]
A01
A02
B01
B02
The playbook
shell> cat pb.yml
- hosts: all
gather_facts: false
tasks:
- block:
- file:
path: newDevices.txt
state: absent
- file:
path: newDevices.txt
state: touch
- lineinfile:
insertafter: EOF
path: newDevices.txt
line: "{{ item }}"
loop: "{{ ansible_play_hosts }}"
delegate_to: localhost
run_once: true
gives
shell> ansible-playbook pb.yml --limit lab[0:1]
PLAY [all] **************************************************************************************
TASK [file] *************************************************************************************
changed: [A01 -> localhost]
TASK [file] *************************************************************************************
changed: [A01 -> localhost]
TASK [lineinfile] *******************************************************************************
changed: [A01 -> localhost] => (item=A01)
changed: [A01 -> localhost] => (item=A02)
PLAY RECAP **************************************************************************************
A01: ok=3 changed=3 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
shell> cat newDevices.txt
A01
A02
The above playbook is not idempotent. The file is always recreated. The result is that the files will always keep the hosts from the last play only.
The simplified playbook below is idempotent. The list of the hosts in the file, however, will be cumulative
shell> cat pb.yml
- hosts: all
gather_facts: false
tasks:
- lineinfile:
create: true
insertafter: EOF
path: newDevices.txt
line: "{{ item }}"
loop: "{{ ansible_play_hosts }}"
delegate_to: localhost
run_once: true
gives
shell> ansible-playbook pb.yml --limit lab[0:1]
PLAY [all] **************************************************************************************
TASK [lineinfile] *******************************************************************************
changed: [A01 -> localhost] => (item=A01)
changed: [A01 -> localhost] => (item=A02)
PLAY RECAP **************************************************************************************
A01: ok=1 changed=1 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
shell> cat newDevices.txt
A01
A02
shell> ansible-playbook pb.yml --limit lab[0:1]
PLAY [all] **************************************************************************************
TASK [lineinfile] *******************************************************************************
ok: [A01 -> localhost] => (item=A01)
ok: [A01 -> localhost] => (item=A02)
PLAY RECAP **************************************************************************************
A01: ok=1 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
shell> ansible-playbook pb.yml --limit lab[0:2]
PLAY [all] **************************************************************************************
TASK [lineinfile] *******************************************************************************
ok: [A01 -> localhost] => (item=A01)
ok: [A01 -> localhost] => (item=A02)
changed: [A01 -> localhost] => (item=B01)
PLAY RECAP **************************************************************************************
A01: ok=1 changed=1 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
shell> cat newDevices.txt
A01
A02
B01
shell> ansible-playbook pb.yml --limit lab[0]
PLAY [all] **************************************************************************************
TASK [lineinfile] *******************************************************************************
ok: [A01 -> localhost] => (item=A01)
PLAY RECAP **************************************************************************************
A01: ok=1 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
shell> cat newDevices.txt
A01
A02
B01
The playbook below is idempotent and the file will keep the list of hosts from the last play only
shell> cat pb.yml
- hosts: all
gather_facts: false
tasks:
- copy:
dest: newDevices.txt
content: |-
{% for host in ansible_play_hosts %}
{{ host }}
{% endfor %}
delegate_to: localhost
run_once: true

Another way to do this would be to have a play targeting localhost itself since the file is to be created on the controller anyway. The file can be created using the "items" of lab inventory group.
Something like below:
- hosts: localhost
connection: local
tasks:
- file:
path: newDevices.txt
state: absent
- lineinfile:
path: newDevices.txt
create: yes
insertafter: EOF
line: "{{ item }}"
loop: "{{ groups['lab'][0:2] }}"
In the above example, the file gets created by lineinfile task itself. You can also remove/change the list slice of [0:2] as per your requirement.
Then run it with:
ansible-playbook myplaybook.yml -i hosts

Related

How Do I Make An Ansible Playbook Skip A Role?

I am trying to skip the upgrade part of my playbook. The key part looks like this:
hosts: linux_group
name: Upgrade the third-party application
roles:
- role: "upgradeEnv"
when: ENV == inventory_hostname
vars:
logdir: "/home/appuser/external/logs"
become_user: "{{ sudoUser }}"
become_method: sudo
become: yes
tags:
- upgrade
And the key part of the role looks like this:
- name: Upgrade database
shell: "upgradeDB.sh {{ env }}"
vars:
DBURL: "{{ user }}#{{ host }}"
no_log: True
register: register_appupgrade
tags:
- upgrade
- fail:
msg: "Upgrade errors:"
when: register_appupgrade.stderr !=""
tags:
- upgrade
I run the playbook with --skip-tags=upgrade but ansible still goes into the role and runs the tasks so I end up with tags: upgrade specified on each task.
The Upgrade database gets skipped but the fail ends the run due to the when condition.
Why is the role being run from the playbook even when the tags are supposed to be skipped?
Why is the fail task not being skipped?
Given the project for testing
shell> tree .
.
├── ansible.cfg
├── hosts
├── pb.yml
└── roles
└── upgradeEnv
└── tasks
└── main.yml
3 directories, 4 files
shell> cat ansible.cfg
[defaults]
inventory = $PWD/hosts
roles_path = $PWD/roles
remote_tmp = ~/.ansible/tmp
retry_files_enabled = false
stdout_callback = yaml
shell> cat hosts
[linux_group]
test_11
test_13
ansible [core 2.14.1]
The tags keyword means the tags are applied to all tasks at the indentation level.
If you apply tags at the play level
shell> cat pb.yml
- hosts: linux_group
roles:
- role: upgradeEnv
when: ENV == inventory_hostname
tags: upgrade
everything will be skipped
shell> ansible-playbook pb.yml --skip-tags=upgrade
PLAY [linux_group] ***************************************************************************
PLAY RECAP ***********************************************************************************
If you apply tags at the role level
shell> cat pb.yml
- hosts: linux_group
roles:
- role: upgradeEnv
when: ENV == inventory_hostname
tags: upgrade
the role will be skipped
shell> ansible-playbook pb.yml --skip-tags=upgrade
PLAY [linux_group] ***************************************************************************
TASK [Gathering Facts] ***********************************************************************
ok: [test_13]
ok: [test_11]
PLAY RECAP ***********************************************************************************
test_11: ok=1 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
test_13: ok=1 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=
You'll see the same result if you omit tags at role level and apply tags at each task in the role
shell> cat pb.yml
- hosts: linux_group
roles:
- role: upgradeEnv
when: ENV == inventory_hostname
shell> cat roles/upgradeEnv/tasks/main.yml
- name: Upgrade database
command: "echo {{ env }}"
register: register_appupgrade
tags: upgrade
- debug:
msg: |
register_appupgrade.stdout: {{ register_appupgrade.stdout }}
register_appupgrade.stderr: {{ register_appupgrade.stderr }}
tags: upgrade
- fail:
msg: Upgrade errors
when: register_appupgrade.stderr != ""
tags: upgrade
If you don't skip tags the play works as expected
shell> ansible-playbook pb.yml -e ENV=test_11 -e env=test
PLAY [linux_group] ***************************************************************************
TASK [Gathering Facts] ***********************************************************************
ok: [test_11]
ok: [test_13]
TASK [upgradeEnv : Upgrade database] *********************************************************
skipping: [test_13]
changed: [test_11]
TASK [upgradeEnv : debug] ********************************************************************
skipping: [test_13]
ok: [test_11] =>
msg: |-
register_appupgrade.stdout: test
register_appupgrade.stderr:
TASK [upgradeEnv : fail] *********************************************************************
skipping: [test_11]
skipping: [test_13]
PLAY RECAP ***********************************************************************************
test_11: ok=3 changed=1 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0
test_13: ok=1 changed=0 unreachable=0 failed=0 skipped=3 rescued=0 ignored=0

How can I call a playbook multiple times?

I have a main login that is as bellow:
- name: Register
hosts: "{{ host }}"
tasks:
- include_role:
name: ROLE_NAME
vars:
service: "{{ service }}"
I want to call my main login multiple time something like bellow:
- name: call main logic
tasks:
- import_playbook:
name: main-logic.yml
loop:
- host: "1"
service: "s1"
- host: "2"
service: "s2"
I didn't find any solution for that, is it possible? HOW?
If you for whatever reason have to run the playbook(s) sequentially create a batch. However, if you want to execute the playbook on all hosts sequentially one after another, use serial. See Setting the batch size with serial. The difference is that you'll see PLAY RECAP once. In the first case, you'll see PLAY RECAP three times.
Create batch
For example,
shell> cat create_batch.yml
- hosts: localhost
vars:
my_batch:
- {host: host1, service: s1}
- {host: host2, service: s2}
- {host: host3, service: s3}
tasks:
- command: which bash
register: which_bash
- copy:
dest: "{{ playbook_dir }}/batch.sh"
mode: "a+x"
content: |
{{ '#!' }}{{ which_bash.stdout }}
{% for i in my_batch %}
ansible-playbook main-logic.yml -e "host={{ i.host }}" -e "service={{ i.service }}"
{% endfor %}
gives
shell> cat batch.sh
#!/usr/bin/bash
ansible-playbook main-logic.yml -e "host=host1" -e "service=s1"
ansible-playbook main-logic.yml -e "host=host2" -e "service=s2"
ansible-playbook main-logic.yml -e "host=host3" -e "service=s3"
Given the project tree for testing
shell> tree .
.
├── ansible.cfg
├── batch.sh
├── create_batch.yml
├── hosts
└── main-logic.yml
0 directories, 5 files
shell> cat ansible.cfg
[defaults]
gathering = explicit
inventory = $PWD/hosts
roles_path = $PWD/roles
remote_tmp = ~/.ansible/tmp
retry_files_enabled = false
stdout_callback = yaml
shell> cat hosts
host1
host2
host3
shell> cat main-logic.yml
- hosts: "{{ host }}"
tasks:
- debug:
msg: "{{ inventory_hostname }} {{ service }}"
The batch gives
shell> ./batch.sh
PLAY [host1] **************************************************************************************
TASK [debug] **************************************************************************************
ok: [host1] =>
msg: host1 s1
PLAY RECAP ****************************************************************************************
host1: ok=1 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
PLAY [host2] **************************************************************************************
TASK [debug] **************************************************************************************
ok: [host2] =>
msg: host2 s2
PLAY RECAP ****************************************************************************************
host2: ok=1 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
PLAY [host3] **************************************************************************************
TASK [debug] **************************************************************************************
ok: [host3] =>
msg: host3 s3
PLAY RECAP ****************************************************************************************
host3: ok=1 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
Setting the batch size with serial
Declare the below variables. For example, in group_vars/all
shell> cat group_vars/all/my_batch.yml
my_batch:
- {host: host1, service: s1}
- {host: host2, service: s2}
- {host: host3, service: s3}
my_batch_host_sevice: "{{ my_batch|
items2dict(key_name='host',
value_name='service') }}"
service: "{{ my_batch_host_sevice[inventory_hostname] }}"
Set serial: 1. The playbook will serve the remote hosts one after another
shell> cat main-logic.yml
- hosts: "{{ host }}"
serial: 1
tasks:
- debug:
var: ansible_play_batch
- debug:
msg: "{{ inventory_hostname }} {{ service }}"
gives
shell> ansible-playbook main-logic.yml -e host='host1:host2:host3'
PLAY [host1:host2:host3] **************************************************************************
TASK [debug] **************************************************************************************
ok: [host1] =>
ansible_play_batch:
- host1
TASK [debug] **************************************************************************************
ok: [host1] =>
msg: host1 s1
PLAY [host1:host2:host3] **************************************************************************
TASK [debug] **************************************************************************************
ok: [host2] =>
ansible_play_batch:
- host2
TASK [debug] **************************************************************************************
ok: [host2] =>
msg: host2 s2
PLAY [host1:host2:host3] **************************************************************************
TASK [debug] **************************************************************************************
ok: [host3] =>
ansible_play_batch:
- host3
TASK [debug] **************************************************************************************
ok: [host3] =>
msg: host3 s3
PLAY RECAP ****************************************************************************************
host1: ok=2 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
host2: ok=2 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
host3: ok=2 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0

Display Ansible playbook with lookups interpolated

I have an Ansible playbook that looks, in part, like this:
...
environment:
F2B_DB_PURGE_AGE: "{{ lookup('env','F2B_DB_PURGE_AGE') }}"
F2B_LOG_LEVEL: "{{ lookup('env','F2B_LOG_LEVEL') }}"
SSMTP_HOST: "{{ lookup('env','SSMTP_HOST') }}"
SSMTP_PORT: "{{ lookup('env','SSMTP_PORT') }}"
SSMTP_TLS: "{{ lookup('env','SSMTP_TLS') }}"
...
Is there any way to run ansible-playbook so that it will show the results of the YAML file after replacing the lookups with their values? That is, I would like to be able to run something like ansible-playbook file.yaml --dry-run and see on standard output (assuming the environment variables were set appropriately):
...
environment:
F2B_DB_PURGE_AGE: "20"
F2B_LOG_LEVEL: "debug"
SSMTP_HOST: "smtp.example.com"
SSMTP_PORT: "487"
SSMTP_TLS: "true"
...
Set the environment for testing
shell> cat env.sh
#!/usr/bin/bash
export F2B_DB_PURGE_AGE="20"
export F2B_LOG_LEVEL="debug"
export SSMTP_HOST="smtp.example.com"
export SSMTP_PORT="487"
export SSMTP_TLS="true"
shell> source env.sh
Given the inventory
shell> cat hosts
localhost ansible_connection=local
Q: "Run something like ansible-playbook file.yaml --dry-run and see environment"
A: The below playbook does the job
shell> cat file.yml
- hosts: all
vars:
my_environment:
F2B_DB_PURGE_AGE: "{{ lookup('env','F2B_DB_PURGE_AGE') }}"
F2B_LOG_LEVEL: "{{ lookup('env','F2B_LOG_LEVEL') }}"
SSMTP_HOST: "{{ lookup('env','SSMTP_HOST') }}"
SSMTP_PORT: "{{ lookup('env','SSMTP_PORT') }}"
SSMTP_TLS: "{{ lookup('env','SSMTP_TLS') }}"
tasks:
- block:
- debug:
msg: |
my_environment:
{{ my_environment|to_nice_yaml|indent(2) }}
- meta: end_play
when: dry_run|d(false)|bool
- debug:
msg: Continue ...
Set dry_run=true
shell> ansible-playbook file.yml -e dry_run=true
PLAY [all] ***********************************************************************************
TASK [debug] *********************************************************************************
ok: [localhost] =>
msg: |-
my_environment:
F2B_DB_PURGE_AGE: '20'
F2B_LOG_LEVEL: debug
SSMTP_HOST: smtp.example.com
SSMTP_PORT: '487'
SSMTP_TLS: 'true'
TASK [meta] **********************************************************************************
PLAY RECAP ***********************************************************************************
localhost: ok=1 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
By default, the playbook will execute tasks
shell> ansible-playbook file.yml
PLAY [all] ***********************************************************************************
TASK [debug] *********************************************************************************
skipping: [localhost]
TASK [meta] **********************************************************************************
skipping: [localhost]
TASK [debug] *********************************************************************************
ok: [localhost] =>
msg: Continue ...
PLAY RECAP ***********************************************************************************
localhost: ok=1 changed=0 unreachable=0 failed=0 skipped=1 rescued=0 ignored=0
Optionally, let the playbook gather facts and use the dictionary ansible_env. Use the filer ansible.utils.keep_keys to select your variables
- hosts: all
gather_facts: true
vars:
my_environment_vars:
- F2B_DB_PURGE_AGE
- F2B_LOG_LEVEL
- SSMTP_HOST
- SSMTP_PORT
- SSMTP_TLS
my_environment: "{{ ansible_env|
ansible.utils.keep_keys(target=my_environment_vars) }}"
tasks:
- block:
- debug:
msg: |
my_environment:
{{ my_environment|to_nice_yaml|indent(2) }}
- meta: end_play
when: dry_run|d(false)|bool
- debug:
msg: Continue ...

Fetch hosts without domain from inventory file

I have following inventory file
$ cat hosts
[web]
server1.example.com
server2.example.com
I would like to fetch the hostname, without the part of domain (.example.com).
I tried with the following playbook, however, it is still fetching with the entire hostname..
$ playbook.yaml
- hosts: localhost
tasks:
- debug:
msg: "{{ groups['web'] }}"
Output
PLAY [localhost] *************************************************************************************************************************************************************
TASK [Gathering Facts] *******************************************************************************************************************************************************
ok: [localhost]
TASK [debug] *****************************************************************************************************************************************************************
ok: [localhost] => {
"msg": [
"server1.example.com"
"server2.example.com"
]
}
PLAY RECAP *******************************************************************************************************************************************************************
localhost : ok=2 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
Expected output
PLAY [localhost] *************************************************************************************************************************************************************
TASK [Gathering Facts] *******************************************************************************************************************************************************
ok: [localhost]
TASK [debug] *****************************************************************************************************************************************************************
ok: [localhost] => {
"msg": [
"server1"
"server2"
]
}
PLAY RECAP *******************************************************************************************************************************************************************
localhost : ok=2 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
You can get what you want from a magic variable called inventory_hostname_short which basically returns anything before the first . found in the inventory_hostname.
To get this in a normal play host loop, it's as easy as:
- hosts: all
tasks:
- name: show target short name
debug:
var: inventory_hostname_short
If you need to get that for hosts not in the host loop, you will have to go through hostvars. Here is an example to get all those names in a list for a given group running from localhost:
- hosts: localhost
gather_facts: false
tasks:
- name: show list of shortnames for group 'toto'
debug:
msg: "{{ groups['toto'] | map('extract', hostvars, 'inventory_hostname_short') }}"
An other example to get that name only for the first server in group 'toto'
- hosts: localhost
gather_facts: false
tasks:
- name: show shortnames for first server in group 'toto'
vars:
server_name: "{{ groups['toto'][0] }}"
debug:
msg: "{{ hostvars[server_name].inventory_hostname_short }}"

Ansible hosts to be set to a substring of a passed variable

I have a play like this:
- name: Perform an action on a Runtime
hosts: all
roles:
- role: mule_action_on_Runtime
A variable at invocation (--extra-vars 'mule_runtime=MuleS01-3.7.3-Testing') has a prefix of the host needed (MuleS01). I want to set hosts: MuleS01. How do I do this?
Given that your pattern is always PartIWant-PartIDonCareAbout-AnotherPartAfterOtherDash you could use the split method of Python, then get the first item of the list via the Jinja filter first.
Here is full working playbook as example:
- hosts: local
gather_facts: no
tasks:
- debug:
msg: "{{ mule_runtime.split('-') | first }}"
This yield the recap:
play.yml --extra-vars 'mule_runtime=MuleS01-3.7.3-Testing'
PLAY [local] *******************************************************************
TASK [debug] *******************************************************************
ok: [local] => {
"msg": "MuleS01"
}
PLAY RECAP *********************************************************************
local : ok=1 changed=0 unreachable=0 failed=0 skipped=0 rescued=0 ignored=0
With the inventory
shell> cat hosts
MuleS01
MuleS02
MuleS03
this playbook
shell> cat pb.yml
- hosts: all
tasks:
- debug:
msg: Set {{ mule_runtime }}
when: mule_runtime.split('-').0 == inventory_hostname
gives
skipping: [MuleS02]
ok: [MuleS01] => {
"msg": "Set MuleS01-3.7.3-Testing"
}
skipping: [MuleS03]

Resources