Snapvault

A PostgreSQL snapshot tool that effortlessly captures and restores precise snapshots of your database.

---

⚠️ Note: This tool is designed for use during development and should not be used in production.

📸 Why Snapvault?

The snapvault CLI tool is intended to be used during local development as an easy way to capture and restore snapshots of the database, making it possible to quickly restore the database to a previous state.

It uses the template functionality in Postgres to create clones of databases, which is faster than using pg_dump/pg_restore. This means that all clones are actually stored as separate databases on the same Postgres server as the original database.

It supports basic commands such as save, restore, list and delete:

$ snapvault save <snapshot_name> 
$ snapvault restore <snapshot_name>
$ snapvault list
$ snapvault delete <snapshot_name>

Snapvault is similar to projects like DLSR and Stellar. However, unlike those projects snapvault is written in Go and delivered as a standalone binary, making it possible to use the tool without having to rely on Python or managing any other dependencies.

⚙️ Installation

Binaries are available in both Intel and ARM versions for OSX/Darwin, Linux and Windows and can be found under the Releases section.

Using Homebrew

$ brew tap cotramarko/tools
$ brew install snapvault

Manual Download

OSX/Darwin & Linux

# Change binary depending on your platform
# Linux: 
#   snapvault_Linux_arm64
#   snapvault_Linux_i386
#   snapvault_Linux_x86_64
# OSX:
#   snapvault_Darwin_arm64
#   snapvault_Darwin_x86_64

# Bash:
$ TARGET=snapvault_Darwin_x86_64
# Fish:
$ set TARGET snapvault_Darwin_x86_64

# Download binary and make it executable
$ sudo curl -fsSL -o /usr/local/bin/snapvault https://github.com/cotramarko/snapvault/releases/latest/download/$TARGET
$ sudo chmod +x /usr/local/bin/snapvault

Windows

Download and unzip the archive for your target platform. The unzipped archive contains an .exe of the binary.

🔧 How to Use Snapvault

Providing Database URL

The database URL can be provided in multiple ways

Option 1 - Using a snapvault.toml file

Create a snapvault.toml file containing the database URL

url = "postgres://acmeuser:acmepassword@localhost:5432/acmedb"

The URL will be used whenever snapvault is in the same directory as the file.

Option 2 - Setting the DATABASE_URL environment variable

Another option is to set the DATABASE_URL environment variable to the database URL

# Bash:
$ export DATABASE_URL=postgres://acmeuser:acmepassword@localhost:5432/acmedb
# Fish:
$ set -x DATABASE_URL postgres://acmeuser:acmepassword@localhost:5432/acmedb

If both the DATABASE_URL is set and a snapvault.toml file is present then the snapvault.toml file will be preferred.

Option 3 - Passing it explicitly with --url flag

Another option is to explicitly pass the database URL with the --url flag whenever running a snapvault command

$ snapvault list --url=postgres://acmeuser:acmepassword@localhost:5432/acmedb

The --url flag will always override any of the other ways of specifying the URL.

Basic Commands

$ snapvault save fix/foobar
Created snapshot fix/foobar

$ snapvault restore fix/foobar
Restored snapshot fix/foobar

$ snapvault list
╭────────────┬──────────────────────┬─────────╮
│ NAME       │        CREATED       │    SIZE │
├────────────┼──────────────────────┼─────────┤
│ fix/foobar │ 2024-06-23T15:37:39Z │ 7561 kB │
╰────────────┴──────────────────────┴─────────╯

$ snapvault delete fix/foobar
Deleted snapshot fix/foobar