Pinventory is an Ansible Dynamic Inventory for locating and categorising Raspberry Pi computers on the local network.
In the Python environment which you are using:
$ python setup.py install
A wrapper script called simply pinventory will be installed
which can be used to invoke the Python package.
To obtain an Ansible inventory of Raspberry Pi computers on the LAN:
$ pinventory --list --pretty
{
"_meta": {
"hostvars": {
"10.0.0.149": {
"hostname": "lime",
"ip": "10.0.0.149",
"mac": "b8:27:eb:a5:93:c5"
},
"10.0.0.186": {
"hostname": "raspberrypi",
"ip": "10.0.0.186",
"mac": "b8:27:eb:73:9d:67"
}
}
},
"raspberries": {
"hosts": [
"10.0.0.149",
"10.0.0.186"
]
}
}
To obtain the host variables for a particular host:
$ pinventory --host 10.0.0.149 --pretty
{
"hostname": "lime",
"ip": "10.0.0.149",
"mac": "b8:27:eb:a5:93:c5"
}
Pass the pinventory executable using the -i inventory option, and
Ansible will execute it as a dynamic inventory. For example, assuming
pinventory is available on the PATH, to run an ad hoc Ansible command
on all Raspberry Pi devices on the LAN, use:
$ ansible raspberries -i `which pinventory` -a <command>
For example, to get the uptime of the Raspberry Pi devices, use:
$ ansible raspberries -i `which pinventory` -a uptime 10.0.0.186 | SUCCESS | rc=0 >> 18:51:14 up 1:03, 1 user, load average: 0.06, 0.02, 0.00 10.0.0.171 | SUCCESS | rc=0 >> 18:51:15 up 1:03, 1 user, load average: 0.03, 0.02, 0.00
By default the returned inventory includes all Raspberry Pi
computers on the local network. The contents of the inventory
and variables associated with each host can be customised by
providing pluggable functions at two pkg_resources entry
points capable of transforming the default inventory and
host variables respectively. Example functions can be found
in the pinventory.transforms.hostsvars and
pinventory.transforms.inventory sub-modules.