Skip to content

Latest commit

 

History

History
51 lines (39 loc) · 1.38 KB

README.md

File metadata and controls

51 lines (39 loc) · 1.38 KB
Raw

Backup role

This role follow a very strict pattern for backup:

  • Run a shell script to generate the backup artifacts and store then in /data/backup
  • Encrypt the backups using the GPG key already installed
  • Upload the encrypted backup to S3
  • Delete all local copies

If something fails, the backup stop as is, so you can manually recover.

The backup folder is cleaned BEFORE the backup, so don't store important things there.

The tasks run after the roles on the demo playbook, keep that in mind.

Variables

  • backup_script - The backup script as text (You can put directly the shell script on the variable)
  • gpg_email - Respective email to use as in (--recipient) for GPG encription
  • bucket - Bucket name for uploading the file
  • folder - Inside the bucket, which folder you will store your files

As obvious, you need your AWS id and secret to perform the uploads to S3, these follow Ansible standards.

Example playbook

- hosts: all
  vars:
    backup_script:
      echo nothing > /data/backup/backup_$(date)
      echo nothing2 > /data/backup/backup2_$(date)
    gpg_email: opsxcq@strm.sh
    bucket: backups
    folder: demo
  tasks:
  - name: My things
    debug: msg="Some additional task here"
  roles:
    - opsxcq.host_backup

Requirements file

- src: git+https://github.com/opsxcq/ansible-role-host-backup.git
  name: "opsxcq.host_backup"