I am using Ansible and I have this configuration in my inventory/all: master 192.168.1.10 ansibleconnection=ssh ansiblesshuser=vagrant ansiblesshpass=vagrant slave 192.168.1.11. A password can be used to authenticate to remote Linux machines, but this is not supported for Mac or OSX systems. The public key must be pre-installed or copied to the remote machine. Installing the extension adds the following Ansible task which runs a given Playbook: When Ansible automation engine is located on an Agent machine.
The basic syntax consists of ansible then the host group from hosts to run against, -m <MODULE_NAME>, and optionally providing arguments via -a 'OPT_ARGS'
It can be controlled via a user's /.ssh/config, via remoteuser in Ansible or through the Ansible inventory. This has changed drastically between Ansible versions pre-2.0 and post 2.0 Follow this link to see how this can be done. How To Run Multiple SSH Command. Run date and hostname commands: $ ssh user@host 'date && hostname' You can run sudo command as follows on a remote box called server1.cyberciti.biz. Next Ansible – Installing Ansible Next Search All Categories ads Ansible Apache Web Server AWS AWS-CSA-Associate c# CentOS Consul Docker git Golang Jenkins Kubernetes Linux Networking OSX Other PostgreSQL PowerShell Puppet Python Rails RedHat RHCE RHCSA rspec-puppet Ruby Rubygems Tutorials Uncategorized Vagrant Vault Virtualbox WIX Wordpress.
Simple Modules
Ping hosts
Osx Enable Vnc
Display gathered facts
Filter gathered facts
Advanced Modules
Copy SSH key manually
Modify file using lineinfile
There are multiple ways to control which user account is used when executing Ansible. It can be controlled via a user's ~/.ssh/config, via remote_user in Ansible or through the Ansible inventory.
This has changed drastically between Ansible versions pre-2.0 and post 2.0
Follow this link to see how this can be done.
When running from the command line, one can just specify which user account to run against directly. Please note that specifying a user can sometime conflict with a user defined in /etc/ansible/hosts
Specifying a user:
Using a specific SSH private key:
Variables can be pesky, but sometimes are required to be passed in via the CLI. Any variable can be set via the command line. Often the command line is the be all, end all in variable overrides.
Passing arguments:
When writing Ansible, sometimes it is tedious to make a change in a playbook or task, then run the playbook It can sometimes be very helpful to run a module directly as shown above, but only against a single development host.
Limit to one or more hosts
This is required when one wants to run a playbook against a host group, but only against one or more members of that group.
Limit to one host
Limit to multiple hosts
Negated limit. NOTE: Single quotes MUST be used to prevent bash interpolation.
Limit to host group
Limiting Tasks with Tags
Limit to all tags matching install
Skip any tag matching sudoers
Busted Cache
Sometimes Ansible has a tendency to hold on to variables too long, which causes Ansible to think that a task/operation had already been done or changed when in fact it didn't.
A simple fix is to flush the redis cache during a code execution.
This can be done like this:
Check for bad syntax
One can check to see if code contains any syntax errors by running the playbook.
Check for bad syntax:
Running a playbook in dry-run mode
Sometimes it can be useful to see what Ansible might do, but without actually changing anything.
One can run in dry-run mode like this:
Modules
Sometimes Ansible just can't cut performing a task using the built-in modules. Raw module to the rescue!
Using raw module to run command similar to running directly via SSH:
Other times, Ansible's modules either aren't well defined yet, or simply do not exist. This is a use case for using shell and command modules. More information for shell and command modules.
The main differences between the two comes down to what kind of command one wishes to run. If the command uses IO redirection of ANY sort, use shell. If the command only contains CLI flags, command module will suffice.
Checking Redis
Command reference: http://redis.io/commands
Sometimes it may be neccessary to manually check Redis for gathered facts from a remote host. Redis is the default back-end for Ansible and usually is running as a server under 127.0.0.1. One can check Redis by performing the following steps:
- Log in to Ansible controller, if Ansible is not running on your local machine
Login to Redis
redis-cliList all stored entries for Ansible
KEYS ansible*Find the entry for the remote host and display the gathered facts for that host
MGET 'ansible_factsXXX.XXX.XXX.XXX'Exit Redis
quit
Manually flushing controller's redis cache
Linux Ansible
Running Redis on OS X as a daemon
Write your own module
There is a wealth of modules available, and writing your own should only be done as an absolute last resort.
If you still wish to write your own, take a look at the docs here: http://docs.ansible.com/ansible/developing_modules.html