An Ansible installation has all necessary parameters for connecting to managed hosts. Duplicating these parameters in ~/.ssh/config is not convenient, especially if a dynamic inventory is used.

This script parses ansible configuration to get these parameters. Just place it somewhere in your $PATH, like ~/.local/bin.

Considered inventory variables: ansible_host, ansible_ssh_host, ansible_user, ansible_ssh_user, ansible_port, ansible_ssh_port, ansible_ssh_private_key_file, ansible_ssh_common_args, ansible_ssh_extra_args.

Considered configuration variables: ssh_args, ssh_executable.

Jinja templates inside variables are rendered, like ansible_ssh_common_args="{{ my_ssh_args }}".

How it works:

user@work$ cd my_ansible_repo
user@work$ cat hosts | grep server1
server1 ansible_host=192.168.0.1 ansible_user=root ansible_port=2222
user@work$ cat ansible.cfg | grep ssh_args
ssh_args = -C -o ControlMaster=auto -o ControlPersist=60s
user@work$ ansible-ssh server1
ssh_args: -C -o ControlMaster=auto -o ControlPersist=60s
ansible_host: 192.168.0.1
ansible_user: root
ansible_port: 2222
ansible_ssh_common_args:
result command: ssh ubuntu@192.168.0.1 -C -o ControlMaster=auto -o ControlPersist=60s -p 2222
root@server1#

Options:

• [-i INVENTORY] specify inventory file for ansible

• all options and arguments after the hostname are passed to ssh.

• Ansible
• jq

Not necessary for end users, just some tooling to use jinja template.

./build.sh

P.S. If you like it, don't forget to put a star on its github repo.