Working with Ansible

Inventory File

This file has the list of hosts where the ansible commands or playbook need to be executed. Default inventory file will be under /etc/ansible/hosts.

Sample Entries

[test-servers]

192.168.0.1

192.168.0.2

192.168.0.3

[appserver]

one.example.com

two.example.com
three.example.com

We can put system name in more than one group, for an instance a server name can be in both test-servers and a appserver group.

Different Port number 

If you have hosts that run on non-standard SSH ports you can mention the port number after the hostname with a colon.

four.example.com:5309

Adding a lot of hosts? 

If you have a lot of hosts following similar patterns you can do this rather than listing each hostname:

[test-servers]

test[01:50].example.com

For numeric patterns, leading zeros can be included or removed, as desired. Ranges are inclusive. You can also define alphabetic ranges:

[appservers]

db-[a:f].example.com

Note: The ‘test-servers‘ in the brackets indicates group names, it is used in differentiate the server deciding which systems you are going to controlling at what times and for what reason.

Now let’s check our all 3 servers by just doing a ping from my localhost (Master Server).

To perform the action we need to use the command ‘ansible‘ with options ‘-m‘ (module) and ‘-all‘ (group of servers).

# ansible -m ping test-servers

Note : If the inventory file is in the default path then we no need to mention the complete path of inventory file. You can specify a different inventory file using the -i <path> option on the command line.

other2.example.com     ansible_connection=ssh        ansible_user=mdehaan

host1 http_port=80 maxRequestsPerChild=808

Testing with Simple modules.

Now we will test with module called ‘command‘, which is used to execute list of commands (like, date, uptime, whoami etc.) on all selected remote hosts at one go, for example watch out few examples shown below.

a)  To check the partitions on all remote hosts

# ansible -m command -a "df -h" test-servers

- m -  Module name

- a  - Command line arguments

b) Check memory usage on all remote hosts.

# ansible -m command -a "free -mt" test-servers

c) Checking Uptime for all 3 servers.

# ansible -m command -a "uptime" test-servers

d) Check for hostname and Architecture.

# ansible -m command -a "arch" test-servers

# ansible -m shell -a "hostname" test-servers

e) If we need the output to any file we can redirect as below.

# ansible -m command -a "uptime" web-servers > /u01/uptime_output.txt

Like this way, we can run many shell commands using ansible as what we have run the above steps.

Executing Commands as other user

By default, Ansible will attempt to remote connect to the machines using your current user name, just like SSH would. To override the remote user name, just use the ‘-u’ parameter.

If you would like to access sudo mode, there are also flags to do that:

# as bruce

$ ansible all -m ping -u bruce

# as bruce, sudoing to root

$ ansible all -m ping -u bruce --sudo

# as bruce, sudoing to batman

$ ansible all -m ping -u bruce --sudo --sudo-user batman

# With latest version of ansible `sudo` is deprecated so use become 

$ ansible all -m ping -u bruce -b

# as bruce, sudoing to batman

$ ansible all -m ping -u bruce -b --become-user batman

(The sudo implementation is changeable in Ansible’s configuration file if you happen to want to use a sudo replacement. Flags passed to sudo (like -H) can also be set there.)

To Run command on all nodes irrespective of your server groups

If we use -all option then ansible will try to ping all the hosts in the inventory file irrespective of the group.

# ansible all -a "/bin/echo hello"

–all Option

# ansible -m ping –all