backup

ssbak = Stupid Simple Backup

Description

This tool allows backing up files to zip files in Linux. An automated process can be setup in order to scan target directories periodically. The process will check which files are not backed up and compress them into new zip files.

Why I made this

The main reason is to have a little side project. The second reason: I wanted to have a very simple tool to backup files in my home server (mainly photos and videos). However, I want the resulting backup files (zip) to be independent from the tool generating them. Someone without a lot of tech knowledge should be able to extract those files anywhere.

Usage

The program uses a CLI. Aside from the information below you can run ssbak --help to get usage info.

ssbak daemon -c <config file> -d <database file> = Service mode

This will run the scheduling service. This requires a configuration file. See below.

The service uses an SQLite database where it keeps track of the backed files (assets) and destination archives.

Config

The configuration file uses JSON format.

Parameters:

• sources: A list of backup sources.
  • source_dir: The source directory. The directory and its subdirectories will be scanned for regular files to backup.
  • archive_dir: The target directory where backup archives will be generated.
  • enable: Whether to schedule this backup.
  • cron: The schedule in UNIX cron format.
  • (optional) archive_prefix: This will be appended to the name of generated archive files.
  • (optional) archive_max_sum_size: The maximum bytes that sum the files being written into archives. This is before compression. Is written in units. Example: "32", "32b", "32K", "32Gb"...
  • (optional) archive_include_large_files: Default is false. Include files greater than archive_max_sum_size even if the compressed archive can end up greater than this size.

Example of minimal config for backup:

{
    "sources": [
        {
            "source_dir": "/backup_target",
            "archive_dir": "/zip_files_dir",
            "enable": true,
            "cron": "0 0 * * *"
        }
    ]
}

Other example:

{
    "sources": [
        {
            "source_dir": "/backup_target_1",
            "archive_dir": "/archives",
            "archive_prefix": "target_1_",
            "enable": true,
            "cron": "0 0 * * *"
        },
        {
            "source_dir": "/backup_target_2",
            "archive_dir": "/archives",
            "archive_prefix": "target_2_",
            "archive_max_sum_size": "32M",
            "archive_include_large_files": true,
            "enable": true,
            "cron": "0 * * * *"
        }
    ]
}

ssbak backup -s <source dir> -D <dest dir> -d <database file> = Manually backup files

This commands scans the source directory for files and copies them into a new archive in target directory. By default only new or modified files are copied. Use the --full flag to backup all files from the source directory.

The files are registered in the database.

ssbak restore -D <restore dir> -d <database file> = Manually restore files

This command will restore files into the target directory. The database is used as a reference source for the files that should be restored.

ssbak clean -d <database file> = Manually clean old files

This command will remove the archives in which all backup files are already backed up in newer archives.

IMPORTANT This will remove previous versions of backup files.

Build

go build -o ssbak

Test

go test -race ./...

Coverage

go test -race -coverprofile=coverage.out -covermode=atomic ./...
go tool cover -html=coverage.out -o coverage.html

Lint

Requires golangci-lint.

golangci-lint run

Container

# Build docker image for compilation.
go build -o ssbak .
# Build ssbak container.
docker build -t ghcr.io/stupid-simple/backup .