ZFShotter

ZFShotter is a utility designed to automate the creation of ZFS snapshots and replicate them to remote servers

Table of Contents

• Features
• Installation
• Configuration
  • Job
  • Datasets
  • Replication
• License

Features

• Automates the creation of ZFS snapshots.
• Flexible prune policy pipeline to automate the pruning of old snapshots.
• Supports replication of snapshots to remote servers (push-based via ssh).
• Easy configuration file for customizable settings.

Installation

To install ZFShotter, follow these steps:

1. Clone the repository to your desired directory (e.g., /opt/, /usr/local/, or wherever you want). In this documentation, we will use /usr/local:

   git clone https://github.com/mojtabamoaddab/zfshotter.git /usr/local/zfshotter

2. Configure Jobs according configuration section.

3. Define a job scheduler (e.g., cron or systemd-timer) to run the jobs with following command:

   /usr/local/zfshotter/bin/zfshotter.sh <JOB_NAME>

Configuration

The configuration for ZFShotter is organized into a config directory, which contains three subdirectories: jobs, datasets, and replications. Each subdirectory holds .conf files that define the respective configurations.

Directory Structure

config/
├── jobs/
│   └── sample.conf
├── datasets/
│   └── sample.conf
└── replications/
    └── sample.conf

Job

A job is a sequence of operations that involves taking snapshots, pruning old ones, and then replicating them.

You can see config/jobs/sample.conf to possible configuration options.

Supported prune-policies (can be combined with |):

Prune Policy	Description	Example
keep_for <duration>	Keep snapshots within specified duration	keep_for 4d12h
keep_n <n>	Keep the last n snapshots	keep_n 20
keep_regex <regex>	Keep snapshots that match the specified regex pattern in their name	keep_regex "^manual"

Datasets

The datasets configuration file is a simple list of datasets (for taking snapshots, pruning, or replicating) along with some per-dataset options. Each line contains a dataset path followed by optional key-value pairs separated by semicolons.

Example:

path/to/dataset; option1=value1; option2=value2

Replication

This section defines the remote server for replicating snapshots and the datasets to replicate.

You can see config/replications/sample.conf to possible configuration options.

License

This project is licensed under the GPL-3.0-or-later License - see the LICENSE file for details.