Giter Site home page Giter Site logo

vaultwarden-backup's Introduction

Overview

Note: Vaultwarden was formerly known as bitwarden_rs.

This repo contains my automated setup for SQLite-based Vaultwarden backups. It's designed solely to meet my own backup requirements (i.e., not to be general purpose):

  1. Generate a single archive with a complete backup of all Vaultwarden data and config on a configurable schedule.

  2. Retain backup archives on the local Vaultwarden host for a configurable number of days.

  3. Upload encrypted copies of the backup archives to one or more cloud storage services using rclone. The retention policy is configured/managed at the storage service level.

  4. Return success when all backup archives are successfully uploaded, or failure if any uploads fail. This allows cron monitoring services like Healthchecks.io, Cronitor, or Dead Man’s Snitch to provide notification of backup failures.

For the most part, I'm not looking for contributions or feature requests, as this repo is only intended to implement my own backup requirements. I may be willing to make some minor generalizations to make it easier for people to use the repo without modification, but aside from that, feel free to fork and modify this setup to fit your own needs.

Note: This single-archive backup scheme isn't space-efficient if your vault includes large file attachments, as they will be re-uploaded with each backup. If this is an issue, you might consider modifying the script to use restic instead.

Prerequisites

  1. A standard Unix-like (preferably Linux) host running Vaultwarden. I don't know much about Synology or other such environments.

  2. A cron daemon. This is used to run backup actions on a scheduled basis.

  3. An sqlite3 binary (https://sqlite.org/cli.html). This is used to back up the SQLite database. This can be installed via the sqlite3 package on Debian/Ubuntu or the sqlite package on RHEL/CentOS/Fedora.

  4. An rclone binary (https://rclone.org/). This is used to copy the backup archives to cloud storage. This can be installed via the rclone package on Debian/Ubuntu and RHEL/CentOS/Fedora (EPEL required for RHEL/CentOS), but as rclone changes more rapidly, it's probably best to just use the latest binary from https://rclone.org/downloads/.

  5. An account at one or more cloud storage services supported by rclone. If you don't have one yet, here are a few cloud storage services that offer a free tier:

  6. Optionally, a gpg (GnuPG 2.x) binary (https://gnupg.org/). This can be installed via the gnupg package on Debian/Ubuntu or the gnupg2 package on RHEL/CentOS/Fedora.

  7. Optionally, an age binary (https://github.com/FiloSottile/age). This option requires a custom version of the tool that supports reading the passphrase from an environment variable.

Usage

  1. Start by cloning this repo to the directory containing your Vaultwarden data directory, under the name backup. In my setup, it looks like this:

    $HOME/vaultwarden  # Top-level Vaultwarden directory
    ├── backup         # This backup repo
    └── data           # Vaultwarden data directory
    
  2. Copy the backup.conf.template file to backup.conf.

    1. If you want encrypted backup archives using gpg, set the GPG_PASSPHRASE variable accordingly. If you want to encrypt using age instead, set the AGE_PASSPHRASE variable. If both variables are set, only gpg encryption will be performed. If you don't want encryption at all, comment out both variables or set them to be blank.

      This passphrase is used to encrypt the backup archives, which may contain somewhat sensitive data in plaintext in config.json (the password entries themselves are already encrypted by Bitwarden). It should be something easy enough for you to remember, but complex enough to deter, for example, any unscrupulous cloud storage personnel who might be snooping around. As this passphrase is stored on disk in plaintext, it definitely should not be your Bitwarden master passphrase or anything similar.

      rclone crypt is another option for encrypted archives. If you prefer to use this method, just set GPG_PASSPHRASE to be blank, configure rclone crypt appropriately, and use the crypt remote in RCLONE_DESTS.

    2. Change RCLONE_DESTS to your list of rclone destinations. You'll have to configure rclone appropriately first.

    3. Note that backup.conf is simply sourced into the backup.sh script, so you can add arbitrary environment variables into backup.conf as needed. This can be useful for configuring any tools called from backup.sh, such as rclone.

  3. Modify the backup/crontab.template file as needed. This crontab actually calls cron.sh to run the backup, rather than calling backup.sh directly. Currently, cron.sh captures the output of the current run of backup.sh to a backup.log file. It also saves a copy of this log file, named according to whether the backup run was a success or failure. You can add other custom logic to cron.sh if needed, such as signaling failure to a cron monitoring service.

    1. If $HOME/vaultwarden isn't your top-level Vaultwarden directory, adjust the paths in this file accordingly.

    2. Review the backup schedule. I generate backup archives hourly, but you might prefer to do this less frequently to save space.

    3. Review the local backup archive retention policy. I delete archives older than 14 days (-mtime +14). Adjust this if needed.

    4. Review the log file retention policy. I delete log files older than 14 days (-mtime +14). Adjust this if needed.

    5. Review the SQLite VACUUM schedule, or remove the job if you don't want vacuuming. Vacuuming compacts the database file so that operations are faster and backups are smaller.

  4. Install the crontab under a user (typically your normal login) that can read your Vaultwarden data. In many cases, running crontab -e and pasting the contents of the filled-in crontab template file should work. Note that if your cron user doesn't have write permissions to the database, then you must ensure it has write permissions to the Vaultwarden data directory, as SQLite may need to create a -wal file for the database if it doesn't already exist. If it's unable to do this, the backup will fail with an attempt to write a readonly database error. (For more details, see https://sqlite.org/wal.html#read_only_databases.)

  5. If you use GnuPG 2.1 or later, see the note about --pinentry-mode loopback in backup.sh.

If everything is working properly, you should see the following:

  1. Backup archives generated under backup/archives.
  2. Encrypted backup archives uploaded to your configured rclone destination(s).
  3. A log of the last backup at backup/backup.log.
  4. Copies of the backup logs saved to backup/logs.

For example:

$HOME/vaultwarden/backup
├── archives
│   ├── vaultwarden-20210101-0000.tar.xz
│   ├── vaultwarden-20210101-0000.tar.xz.gpg
│   ├── vaultwarden-20210101-0100.tar.xz
│   ├── vaultwarden-20210101-0100.tar.xz.gpg
│   └── ...
├── backup.conf
├── backup.conf.template
├── backup.log
├── backup.sh
├── cron.sh
├── crontab.template
├── logs
│   ├── backup-success-20210101-0000.log
│   ├── backup-success-20210101-0100.log
│   ├── backup-failure-20210101-0200.log
│   └── ...
├── LICENSE
└── README.md

vaultwarden-backup's People

Contributors

jjlin avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.