Application Deployment with Rollback Capability

Problem/Feature Description

A fintech company deploys their Java billing service to a group called app_servers. Deployments occasionally fail mid-way (e.g., the new JAR file is copied but the service fails to start), leaving servers in a broken state. The on-call team then spends 30 minutes manually rolling back.

The deployment process is:

1. Stop the application service
2. Back up the current JAR at /opt/billing/billing.jar to /opt/billing/billing.jar.bak
3. Copy the new JAR from the control node to /opt/billing/billing.jar
4. Start the service and wait for it to listen on port 8080

If any step from (3) onwards fails, the playbook should automatically restore the backup JAR and restart the service. Regardless of success or failure, the deployment status (success or failure with the error message) should be written to /var/log/ansible-deploy.log.

Output Specification

Produce a complete Ansible playbook file named deploy_billing.yml that implements the deployment with automatic rollback.

Input Files

The following existing playbook stub is provided as a starting point. Extract it before beginning.

=============== FILE: deploy_billing_stub.yml ===============

• name: Deploy billing service hosts: app_servers become: yes

  vars: billing_jar_src: "files/billing.jar" billing_jar_dest: "/opt/billing/billing.jar" billing_service_port: 8080

  tasks:

  • name: stop billing service ansible.builtin.service: name: billing state: stopped ignore_errors: yes

  • name: backup current jar command: cp /opt/billing/billing.jar /opt/billing/billing.jar.bak

  • name: copy new jar copy: src: "{{ billing_jar_src }}" dest: "{{ billing_jar_dest }}" ignore_errors: yes

  • name: start service service: name: billing state: started ignore_errors: yes