DRLM (Disaster Recovery Linux Manager)Dormant

A central server for Relax-and-Recover (ReaR): it runs whole-system backups of many Linux machines over the network, keeps each one as a disk image, and boots a broken machine over PXE straight into a recovery system that restores it.

What it is: Backup & restore
Type: Open source
Made by: Brain Updaters, S.L.
License: GPL-3.0
Status: Dormant. The last release, 2.4.13, came out in April 2025, and the public repository has had no commits since May 2025.
Facts checked: 24 October 2026

What it is

Relax-and-Recover (ReaR) is a bare-metal disaster recovery tool for Linux: on each machine it builds a small rescue system plus a backup of the files, and if the disk dies you boot the rescue system and run rear recover to rebuild partitions, file systems and data. DRLM puts a server in front of many ReaR clients. You register a client with drlm addclient, which can install and configure ReaR on it over SSH, then run drlm runbackup or schedule jobs, and the server collects every client’s rescue system and backup.

Each backup is stored on the server as a DR file, a QCOW2 disk image. To make a backup available, DRLM attaches the image to a network block device with qemu-nbd, mounts it read-only and exports it over NFS or an rsync module; a SQLite database records clients, networks, jobs and which backup is enabled. The server also runs DHCP and TFTP, so a failed client set to PXE can boot from the network into ReaR’s recovery system and restore from the enabled backup. Clients can instead get an ISO image, and data-only backups can restore single files back to a running machine.

DRLM is written in Bash, like ReaR, with an API service in Go. Server packages cover Debian, Ubuntu, Rocky Linux, RHEL, CentOS and SUSE. Brain Updaters, the Spanish company behind it, sells support and an enterprise edition with extras such as malware scanning after a backup.

Why teams pick it

Whole-machine recovery, not just files: a dead server boots from the network and comes back with its partitions, boot loader and data
One server manages ReaR for many clients, with central scheduling, backup policies and error reports to Nagios, Zabbix, mail or Telegram
Backups are disk images the server can mount, export and switch between, so you choose which restore point a client boots into
Useful for migrations as well as recovery: a backup can be restored onto new hardware or a virtual machine

Watch out for

Development appears to have stopped: no release since 2.4.13 in April 2025 and no commits since May 2025
Linux only, and the server wants a dedicated machine because it takes over DHCP, TFTP, NFS and rsync
Some features, such as scanning backups for malware, archiving and syncing, are reserved for the commercial enterprise edition
Not a database backup tool: ReaR copies file systems with tools such as tar or rsync, so databases still need their own consistent dumps or archived logs

Alternatives

What all backup & restore have in common →Compare side by side →

BaculaOpen sourceA client-server backup system for networks of mixed machines: a central Director schedules jobs, agents on each machine send files to storage daemons that write disk or tape volumes, and a SQL catalog records where every file went.

DuplicatiOpen sourceA backup client for Windows, macOS and Linux that splits files into blocks, stores each block once, and uploads compressed, encrypted volumes to cloud storage or a file server, driven from a web UI or the command line.

pgBackRestOpen sourceA PostgreSQL-only backup tool in C: parallel full, differential and incremental backups, WAL archiving and point-in-time restores, to disk, S3, Azure, GCS or SFTP.

PlakarOpen sourceA backup engine in Go that takes deduplicated, encrypted, immutable snapshots of files, databases and cloud sources into a store called a Kloset, which you can browse, diff, verify and sync without restoring.

VeleroOpen sourceBacks up a Kubernetes cluster’s objects and persistent volumes to object storage, and restores them into the same cluster or a new one.

WAL-GOpen sourceA single Go binary that streams compressed, encrypted Postgres backups and WAL to object storage, and restores them in parallel.

How it fits with other technologies

What DRLM (Disaster Recovery Linux Manager) uses

Add a client, back it up, and see which backups the server holds · bash

# on the DRLM server: register the client and install ReaR on it over SSH
drlm -v addclient -i 192.168.181.53 -c ReaRCli1 -I

# take a full-system backup now
drlm -v runbackup -c ReaRCli1

drlm listclient
drlm listbackup -c ReaRCli1

# to recover: boot ReaRCli1 from the network, pick the recovery entry,
# log in as root and run
rear recover

See it in a real stack

No architecture uses DRLM (Disaster Recovery Linux Manager) yet. Drop it into the playground and wire it up yourself.Open playground →

Where it came from2006–2025

ReaR itself dates from 2006, when Gratien D’haese and Schlomo Schapiro rewrote the older mkCDrec rescue tool; by mid-2015 it was part of Red Hat Enterprise Linux. ReaR is set up and run on each machine separately. Brain Updaters, a Spanish company, started DRLM on GitHub in October 2013 to manage it centrally; an early write-up says it was first built to solve a problem that hits large data centres, and only later used more widely.

DRLM 2.0 came out in July 2016, and at FOSDEM 2017 Didac Oliveira ran a workshop on disaster recovery management with ReaR and DRLM, covering version 2.1. Version 2.3 in 2019 replaced the Apache and CGI web layer with an API written in Go. Version 2.4 in October 2021 changed the DR file format from raw images on loop devices to QCOW2 images on network block devices, and added incremental backups, ISO recovery images, PowerPC support, rsync transport and encrypted backup files.

The 2.4.x line continued with Ubuntu 24.04 support, backup policies, TLS for rsync and a file-restore workflow, and 2.4.12 in February 2025 laid the base for enterprise features. The last release, 2.4.13 in April 2025, added support for ReaR 2.9; the public repository has been quiet since May 2025.

Why it’s built this way

Manage ReaR instead of writing a new backup tool: ReaR already knew how to build a rescue system and rebuild a Linux machine from scratch. DRLM adds what one machine can’t do alone: a central catalogue, a schedule and a network boot server.
Recover by network boot: Serving the rescue system over PXE and TFTP means a machine with an empty or replaced disk can be rebuilt without anyone carrying a USB stick or ISO to it.
Backups as mountable disk images: Keeping each backup as an image the server attaches and exports read-only lets an administrator switch which restore point a client sees by enabling one image and disabling another.

Timeline

2006
Relax-and-Recover (ReaR) created from a rewrite of mkCDrec
2013
Brain Updaters starts the DRLM repository on GitHub
2016
DRLM 2.0 in July
2017
DRLM workshop at FOSDEM; version 2.1
2019
DRLM 2.3 replaces the CGI web layer with a Go API
2021
DRLM 2.4: QCOW2 images on network block devices, incremental backups, ISO recovery
2025
Last release, 2.4.13, in April

Sources8 · facts checked 24 October 2026

Improve this page on GitHub