Closed. This question is not reproducible or was caused by typos. It is not currently accepting answers.
This question was caused by a typo or a problem that can no longer be reproduced. While similar questions may be on-topic here, this one was resolved in a way less likely to help future readers.
Closed 4 years ago.
Improve this question
All,
My playbook is almost ready. Only the last line with alias is giving me issues. I tried to look at what I am doing wrong in the yaml syntax checker but still not able to find it. Any clue what I am doing wrong? I tried escaping the forward slash but still no luck :(
I am bad at finding the syntax errors in general. So appreciate any clues you folks can provide in finding the problem.
tasks:
- name: Create a KMS key for using the aws cli
command: 'aws kms create-key --profile="{{ aws_profile }}" --region="{{ aws_region }}"'
register: newkeydetails
- name: Display the values for the variable output
set_fact: newkeydetails="{{ newkeydetails.stdout | from_json }}"
- name: Display the value of keyid
debug:
msg: "{{ newkeydetails.KeyMetadata.KeyId }}"
- name: Create a alias name for KMS key using the aws cli
command: 'aws kms create-alias --alias-name 'alias/anothernewkeydetailskey' --target-key-id '"{{ newkeydetails.KeyMetadata.KeyId }}"' --profile="{{ aws_profile }}" --region="{{ aws_region }}"'
The reason for the issue here was due to the usage of single quotes in the last line, as they were not escaped, resulting in parsing errors.
The working solution involves removing unnecessary quotes, resulting in the following:
'aws kms create-alias --alias-name "alias/anothernewkeydetailskey" --target-key-id "{{ newkeydetails.KeyMetadata.KeyId }}" --profile="{{ aws_profile }}" --region="{{ aws_region }}"
Related
Closed. This question is not reproducible or was caused by typos. It is not currently accepting answers.
This question was caused by a typo or a problem that can no longer be reproduced. While similar questions may be on-topic here, this one was resolved in a way less likely to help future readers.
Closed 4 months ago.
Improve this question
I try to run task depending on two conditions using logical "and". For the first condition I created the following task:
- name: Check if configuration already exists, then skip next task
stat:
path: "{{openldap_config}}/cn=config"
register: is_configured
Then I crated the following task, which is working fine:
- name: add new basic-configuration
shell: /opt/symas/sbin/slapadd -n 0 -F {{openldap_config}} -l /home/{{ansible_user}}/config.ldif
args:
executable: /bin/bash
when: not is_configured.stat.exists
If a configuration file exists the task will be skipped. Then I have a few task where I use "group_names" like this on:
- name: generating deltasyncrepl LDIF for main DB
template:
src: main_db_repl.j2
dest: /home/{{ansible_user}}/main-db-repl.ldif
owner: "{{ansible_user}}"
group: "{{ansible_group}}"
mode: '660'
when: "'ldap_provider' in group_names"
That's also working. BUT now I would like to have a logical "and" for both conditions:
- name: add replication of cn=config to all provider
shell: /opt/symas/bin/ldapmodify -Y EXTERNAL -H ldapi:/// -f /home/{{ansible_user}}/repl_config.ldif
args:
executable: /bin/bash
when: not is_configured.stat.exists
and "'ldap_provider' in group_names"
This is not working. I tried different quoting but I could not find a a working solution. I try it with brackets too, also not working. Some how it must be possible to have a logical AND with a int- and a string- variable.
Thanks for any help
While you can manually stat the file and that works just fine, this is actually exactly the case that the creates parameter to the shell and command modules is designed to handle. You should also migrate to the cmd parameter (supported since Ansible 2.0) instead of using args, so that people reading your code don't have to understand that rarely-used syntax:
- name: Basic configuration for OpenLDAP
shell:
cmd: /opt/symas/sbin/slapadd -n 0 -F {{ openldap_config }} -l /home/{{ ansible_user }}/config.ldif
executable: /bin/bash
creates: "{{ openldap_config }}/cn=config"
register: openldap_config_result
While you didn't include your other attempts at solving this problem, generally the best way to apply multiple conditions to a task is to use a list, which will implicitly apply and between each condition in the list:
- name: Add replication of cn=config to all providers
shell:
cmd: /opt/symas/bin/ldapmodify -Y EXTERNAL -H ldapi:/// -f /home/{{ ansible_user }}/repl_config.ldif
executable: /bin/bash
when:
- openldap_config_result is changed
- "'ldap_provider' in group_names"
You can also do it as a single string, though you have to make sure your quoting is correct (which it isn't in your question):
when: openldap_config_result is changed
and 'ldap_provider' in group_names
Closed. This question is not reproducible or was caused by typos. It is not currently accepting answers.
This question was caused by a typo or a problem that can no longer be reproduced. While similar questions may be on-topic here, this one was resolved in a way less likely to help future readers.
Closed 8 months ago.
Improve this question
I try to subtract 2 variables in Ansible, but what I get is zero in the result.
- name: ansible calculate space
hosts: pghosts
gather_facts: false
tasks:
- name: Check /srv freespace
shell: df /srv --output\=avail | tail -1
register: srv_freespace
- name: Check postgres data dir size
shell: du -ks /srv/postgresql/data |awk '{ print $1 }'
register: postgres_data
- debug:
msg: "substraction {{ srv_freespace|int - postgres_data|int }}"
- copy: content="{{ srv_freespace|int - postgres_data|int }}" dest=/tmp/results.txt
Try changing the last line to:
- copy: content="{{ srv_freespace.stdout|int - postgres_data.stdout|int }}" dest=/tmp/results.txt
Explanation: the registered variable is not a string, its a dict. The dict includes several keys like the command name, command start and end time, exit code and other good stuff. The actual value i.e. what is printed to standard output is under the stdout key.
Closed. This question is not reproducible or was caused by typos. It is not currently accepting answers.
This question was caused by a typo or a problem that can no longer be reproduced. While similar questions may be on-topic here, this one was resolved in a way less likely to help future readers.
Closed 1 year ago.
Improve this question
When a registered variable has a single quote in it, for example
if value of var.stdout is IxMy5NPG'dCbrHD (note the single quote in string)
When I use this variable in an upcoming task,
lineinfile:
path: /path/to/file.j2
regexp: "X1234"
line: {{ var.stdout }}
mode: 0755
state: present
I get the follwing error
ERROR! failed at splitting arguments, either an unbalanced
jinja2 block or quotes
This only occurs if there is a single quote in the string, which may or may not have a single quote. Hence I cannot expect this already and handle it. I tried a regex_replace filter on the variable like {{ var.stdout | regex_replace("'", "\'") }} I get "Unexpected failure during module execution."
The expression {{ var.stdout }} must be quoted
lineinfile:
path: file.j2
regexp: X1234
line: {{ var.stdout }} # <-- WRONG
For example, given the file
shell> cat file.j2
X1234
The task below
lineinfile:
path: file.j2
regexp: X1234
line: "{{ var.stdout }}" # <-- CORRECT
works as expected
shell> cat file.j2
IxMy5NPG'dCbrHD
If you want to replace the quote for example with the underscore "_"
- set_fact:
var_stdout: "{{ var.stdout|regex_replace(_regex, _replace) }}"
vars:
_regex: "'"
_replace: "_"
gives
var_stdout: IxMy5NPG_dCbrHD
To avoid escaping put _regex/_replace into the task's vars.
If you want to escape the quote
_replace: "\\'"
gives
var_stdout: IxMy5NPG\'dCbrHD
Closed. This question is not reproducible or was caused by typos. It is not currently accepting answers.
This question was caused by a typo or a problem that can no longer be reproduced. While similar questions may be on-topic here, this one was resolved in a way less likely to help future readers.
Closed 1 year ago.
Improve this question
I have an ansible task that writes 2 lines into journald.conf, however it doesn't perform idempotency when running again.
I have seen the following questions that didn't work for me:
Ansible lineinfile duplicates line
Idempotency in ansible playbook
https://github.com/ansible/ansible/issues/4531
My regex seems to be okay, you can see below my task:
- name: set cpu affinity settings in systemd
lineinfile:
dest: /etc/systemd/journald.conf
line: "{{ item.key }}={{ item.value }}"
regexp: "^#?{{ item.value }}"
state: present
with_dict:
RateLimitIntervalSec: 0
RateLimitBurst: 0
tags: journald
notify: restart journald
The expected behavior should be: keep the commented line and add new ones at the end of the file with the items in the list, unless the uncommented lines already exists.
My file journald.conf file is like this:
[Journal]
#Storage=auto
#Compress=yes
#Seal=yes
#SplitMode=uid
#SyncIntervalSec=5m
#RateLimitIntervalSec=30s
#RateLimitBurst=1000
#SystemMaxUse=
#SystemKeepFree=
#SystemMaxFileSize=
#SystemMaxFiles=100
#RuntimeMaxUse=
#RuntimeKeepFree=
#RuntimeMaxFileSize=
#RuntimeMaxFiles=100
#MaxRetentionSec=
#MaxFileSec=1month
#ForwardToSyslog=yes
#ForwardToKMsg=no
#ForwardToConsole=no
#ForwardToWall=yes
#TTYPath=/dev/console
#MaxLevelStore=debug
#MaxLevelSyslog=debug
#MaxLevelKMsg=notice
#MaxLevelConsole=info
#MaxLevelWall=emerg
#LineMax=48K
RateLimitIntervalSec=0
RateLimitBurst=0
RateLimitIntervalSec=0
RateLimitBurst=0
RateLimitIntervalSec=0
RateLimitBurst=0
RateLimitIntervalSec=0
RateLimitBurst=0
I tried to use the parameter backrefs: yes as suggested in above mentioned articles, but it performs idempotency every time, even when there is no any uncommented line.
Do you guys have any suggestions?
I'm using ansible 2.9.0
I'd suggest an alternative approach - use the ini_file module, as settings in journald.conf are INI style key=value (with a section as well). This will simplify the task required and be idempotent as well.
Example:
- name: set cpu affinity settings in systemd
ini_file:
path: /etc/systemd/journald.conf
section: Journal
option: "{{ item.key }}"
value: "{{ item.value }}"
no_extra_spaces: yes
with_dict:
RateLimitIntervalSec: 0
RateLimitBurst: 0
Note: If you want a reference of settings prior to change, add backup: yes to the task.
This question already has answers here:
How to use Ansible's with_item with a variable?
(2 answers)
Closed 5 years ago.
I'm using ansible 2.1.0. I'm looking at this page http://ryaneschinger.com/blog/securing-a-server-with-ansible/ and running the playbook of the following part:
- name: Add authorized keys for deploy user
authorized_key: user={{ username }}
key="{{ lookup('file', item) }}"
with_items: public_keys
When I run this, I get [DEPRECATION WARNING]: Using bare variables is deprecated. Update your playbooks so that the environment value uses the full variable syntax ('{{public_keys}}'). What is the not deprecated way to do the same?
As stated in the documentation, bare variables in with_ loops should instead use the “{{var}}” syntax, which helps eliminate ambiguity.
So, It is just telling you to change this:
with_items: public_keys
to this:
with_items: "{{ public_keys }}"