Understanding the Ansible Inventory File When Managing Junos Devices

Inventory File Overview

The Ansible inventory file defines the hosts and groups of hosts upon which commands, modules, and tasks in a playbook operate. The file can be in one of many formats depending on your Ansible environment and plugins. Common formats include INI and YAML.

The default location for the inventory file is /etc/ansible/hosts. You can also create project-specific inventory files in alternate locations, for example, in your project's playbook directory. Your Ansible configuration file can specify the path to a non-default inventory file location. You can also specify the location of the inventory file when you execute a playbook.

The inventory file can list individual hosts or user-defined groups of hosts. Thus, you can organize Junos devices with similar roles into groups. Then you can perform the same operational and configuration tasks on the devices in that group. For example, if you are managing one or more data centers, you can create groups for those switches that require the same set of operations, such as applying similar configurations and upgrading Junos OS.

In order to manage Junos devices using Ansible, you must have a Junos OS login account with appropriate access privileges on each managed device. You must ensure that usernames and passwords or access keys exist for each host in the inventory file.

The following INI-formatted sample inventory file defines an individual host, host1, and two groups of hosts, routers and switches. The inventory also defines a parent group network that includes the routers and switches groups.

For more information about the Ansible inventory file, see the official Ansible documentation at https://docs.ansible.com/projects/ansible/latest/inventory_guide/intro_inventory.html.

juniper.device Collection: Connection and Network OS Variables

Starting in juniper.device Release 2.0.0, we consolidated the Juniper Networks juniper.device collection and the Red Hat Ansible junipernetworks.junos collection into a single juniper.device collection. Although you execute the modules under the same namespace, each module set uses different Ansible connection and network OS values, so you cannot intermix the modules in the same playbook.

You can define the Ansible connection in your playbook, your inventory file, in host or group variable files, or in command-line arguments. If you define the connection in the inventory file, set ansible_connection to the appropriate value for your modules. Table 1 outlines the connection values for each module set. For the junipernetworks.junos modules, you must also define ansible_network_os.

Table 1: Ansbile Connection and Network OS Parameters

Module Set	ansible_connection Value	ansible_network_os Value
juniper.device   (original modules)	local juniper.device.pyez	–
junipernetworks.junos   (all junos_* modules except   junos_netconf and junos_ping)	ansible.netcommon.netconf	juniper.device.junos
junipernetworks.junos   (junos_netconf and junos_ping)	ansible.netcommon.network_cli	juniper.device.junos

Note:

Instead of defining ansible_connection in the inventory file, you can also define connection in your playbook's play. Both parameters take the same values.

In addition to the required connection and network OS variables, your inventory file can optionally define the username that Ansible uses to connect to the devices. It can also specify a port, if it is different from the default port. We recommend that you also define the Python interpreter path in either the Ansible configuration file or the Ansible inventory file. Specifying the Python interpreter path prevents unnecessary errors if Ansible cannot locate the Python installation that your environment uses.

The following sections provide some simplified example inventory files for each module set.

• Sample Inventory File for the juniper.device Modules
• Sample Inventory File for the junipernetworks.junos Modules