How to Use Ansible to Deploy Applications on Remote Servers

Objectives

1. Use ansible to deploy applications on remote servers.

Prerequisites

1. A VPN account.
2. A VPN client installed on your computer.
3. An ssh client installed on your computer.
4. An ssh authentication agent running on your computer.
5. An account for the Device Management application.
6. An OS account on "fcadminprod02.fullcount.net".
7. An ssh "authorized_keys" file set up for your OS account on fcadminprod02.fullcount.net.
8. The "Production Servers - Ed25519" and "Remote Servers" ssh keys loaded into your ssh authentication agent.

Instructions

1. Start the VPN client on your computer and log into the VPN.
2. Load the "Central Servers" and "Remote Servers" ssh key into the authentication agent on your computer.
3. Log into "fcadminprod02.fullcount.net" via ssh.
4. Change your directory to the directory where the "check_remote_connections.yml" playbook is located:
   
   # cd /srv/ansible/playbooks/groups/remote_servers
    
5. Run the "check_remote_connections.yml" playbook to check the status of connections to remote servers:
   
   # ansible-playbook check_remote_connections.yml
    
6. Request connections to any remote servers that have a connection status of FAILED or, that do not have a connection. You can use the Device Management application to do this, or edit and run the "request_connections_to_remote_servers.sql" script.
7. At 10:45pm CST, run the "deployment.yml" playbook for one remote server:
   
   # ansible-playbook deployment.yml -l @do-first
    
8. Ask the Development team to check the results.
9. Be sure to update the skip-list file with those communities that should be skipped for the deployment.  At present this includes the community from step 7. above along with any communities that may have migrated from on-premise to cloud-hosted and should be skipped (for varying reasons).
• Also keep in mind the --list-hosts flag may be used on any of the below commands to verify the target list before executing the deployment.
10. At 11:00pm CST, run the "deployment-all.yml" playbook for the remaining remote servers:
    
    # ansible-playbook deployment-all.yml -l @skip-list
     
11. Copy and paste the serial numbers of any VMs where the deployment failed into the text file redo-list.
12. Correct the problem(s), and try the deployment again:
    
    # ansible-playbook deployment-all.yml -l @redo-list