Website review
This review may include AI-generated content and is not individually verified. Domain, certificate and other facts come from the observed data on this page.
What is SystemRescue?
SystemRescue is a bootable Linux toolkit for repairing and administering a computer after a crash or before an OS reinstall. You boot it from a USB stick or CD/DVD, so it runs independently of the installed operating system, and it works on both Linux and Windows machines, desktops and servers. It includes partitioning and filesystem tools such as GParted, GNU Parted and FSArchiver, plus editors, Midnight Commander and network tools, and its kernel supports ext4, xfs, btrfs, vfat and ntfs as well as Samba and NFS.
The practical use is recovery and maintenance from outside the broken system: resizing or repairing partitions, mounting an NTFS disk read-write, backing up data from an unbootable Windows PC, or transferring files over the network with rsync. It requires no installation to run, though it can be installed to a hard disk if you prefer, and it can also boot over PXE for network-based use.
A concrete scenario: a Windows laptop no longer starts. You write SystemRescue to a USB stick, boot from it, mount the internal drive, copy important files to an external disk, then use GParted to inspect or repair the partition layout before reinstalling Windows.
As a next step, read the Quick Start Guide on SystemRescue before your first boot, since boot options and network setup are the parts most likely to slow down a beginner.
How do I create a bootable USB stick with SystemRescue?
SystemRescue is designed to boot from a USB stick, and the project documents this as a standard setup path. The official homepage links a dedicated guide, "How to install SystemRescue on an USB-stick," alongside its Quick Start Guide, so that page is the place to follow exact steps for your version.
The practical workflow
- Download the ISO from SystemRescue and, if offered, verify its checksum so a corrupted download doesn't fail mid-repair.
- Write the ISO to the USB stick in raw image mode, not by copying the file onto a formatted drive. The stick will be erased, so use one with nothing you need.
- Boot from the stick via your firmware boot menu (often F12, F11 or Esc at power-on). You may need to disable Secure Boot or select the appropriate boot entry; the project's boot-options documentation covers this.
- Read the Quick Start Guide first if this is your first use—it explains the live environment, networking and mounting disks before you touch partitions.
Choosing a writing tool
| Situation | Sensible approach |
|---|---|
| You already run Linux | dd or a graphical writer included with your distribution |
| You work from Windows | A raw-image USB writer such as Rufus or balenaEtcher |
| You want a repeatable, documented method | Follow the project's own USB-stick guide step by step |
A common reader scenario: a Windows laptop won't boot and you want to copy files off before reinstalling. You write the ISO from a working machine, boot the failing laptop from the stick, mount the NTFS partition read-write, and transfer data to an external drive—the homepage explicitly covers both backing up data from an unbootable Windows computer and mounting NTFS with full read-write support.
As a next step, open the USB-stick installation page on the official site and match its instructions to your operating system before writing anything.
Can SystemRescue recover data from a Windows computer that won't boot?
Yes. SystemRescue is explicitly designed for repairing systems and data after a crash, and it can be booted from a USB stick or CD/DVD, so it does not depend on the broken Windows installation starting. The project's own documentation lists "Backup data from an unbootable Windows computer" as a supported task.
How it applies to a non-booting Windows PC
- Boot from removable media. The rescue environment runs in memory from the USB stick or disc, so the failing Windows drive is only read, not required to boot.
- Read Windows filesystems. The kernel supports NTFS and vfat, so internal drives and USB disks formatted by Windows can be mounted and browsed. The documentation covers mounting NTFS with full read-write support.
- Copy data off. You can move files to an external USB drive, or over the network using Samba or NFS shares; the included rsync guide covers backup and transfer.
- Check the disk itself. GParted and GNU Parted let you inspect partitions, and the included filesystem tools help determine whether the problem is a damaged partition table, a full disk, or a failing drive.
- Repair boot problems. The site's guides cover how GRUB boots and how to repair it, which is often the actual cause of a Windows machine that will not start.
A practical scenario
A laptop shows a boot error after a failed update. You write SystemRescue to a USB stick, boot it, mount the NTFS partition, copy the user's documents and photos to an external drive, then run a filesystem check. If the data is intact and only the bootloader is broken, the GRUB repair guide may get Windows starting again; if the drive is failing, you have already secured the files.
Decision criteria
- Choose SystemRescue when you need a general-purpose Linux rescue toolkit with partitioning, archiving and network tools in one bootable image, and you are comfortable working from a command line or a partition editor.
- Consider a Windows-focused alternative if you prefer a graphical file-copy tool built around Windows filesystems, or if you need to recover deleted files rather than copy existing ones. For example, Clonezilla is a well-known option for disk imaging and cloning when the goal is a full backup before repair.
Next step
Read the Quick Start Guide first, then the chapters on creating the USB stick and on backing up data from an unbootable Windows computer. Work on a copy where possible, and avoid writing to the failing drive until your data is safely elsewhere.
How do I use GParted in SystemRescue to resize or repair a partition?
Boot SystemRescue from its USB or CD/DVD medium, open the desktop menu, and launch GParted from the system tools. GParted is the GUI front end to GNU Parted included in SystemRescue, so you can select a disk, review its partition layout, and then choose the partition you want to resize or repair. The page evidence lists GParted among the included tools and describes the toolkit as intended for administering or repairing a system and its data after a crash, on both Linux and Windows computers.
Resizing a partition
- Start SystemRescue and let it reach the desktop.
- Open GParted and pick the correct disk from the drop-down at the top right.
- Select the target partition and choose Resize/Move.
- Enter the new size or drag the partition edges, then confirm.
- Apply the pending operations with the checkmark button and wait for the job to finish.
For shrinking, back up important data first, since a resize can fail or leave a filesystem inconsistent. GParted cannot resize a partition that is mounted with pending changes, and the page evidence notes that SystemRescue can mount NTFS partitions with full read-write support, so be deliberate about which volumes you mount before editing.
Repairing a partition
GParted is mainly a partition editor; filesystem repair is usually done with the filesystem-specific tools that SystemRescue also bundles, such as the filesystem utilities mentioned on the homepage. A practical sequence is:
- Use GParted to check the partition table and confirm the partition exists and has the expected type.
- Unmount the partition in GParted before running any check.
- Run the appropriate filesystem check tool for ext4, xfs, btrfs, vfat or ntfs, as listed in SystemRescue's included tools.
- Return to GParted to verify the partition layout is intact.
A concrete scenario
A reader with a Windows laptop that no longer boots can start SystemRescue from USB, open GParted to confirm the NTFS partition is present, then mount it read-write only when needed to copy files off, following the guide on mounting NTFS with full read-write support. If the partition table itself looks wrong, the disk partitioning guides on SystemRescue cover GPT disks and Grub repair.
The main trade-off is that GParted is convenient for layout changes but is not a substitute for a filesystem-specific repair tool; use it for resizing and partition-level fixes, and use the dedicated utilities for filesystem checks. As a next step, read SystemRescue's Quick Start Guide before your first rescue session so you know the boot options and how to reach GParted.
How can I save my changes or custom software between SystemRescue sessions?
SystemRescue is a live toolkit, so changes normally disappear on reboot. To keep them, you either create a persistent backing store or install the system to disk. The project's own documentation describes both approaches, and the choice depends on whether you want a portable rescue stick or a semi-permanent installed environment.
Two main options
| Approach | What it does | Best for | Trade-off |
|---|---|---|---|
| Backing store | A file that saves your modifications between boots of the live medium | Keeping settings, scripts or added packages on a USB stick | Tied to that medium; needs to be set up and managed |
| Install to disk | Writes SystemRescue onto a hard disk so it behaves like an installed system | A dedicated rescue partition or machine you use repeatedly | Uses disk space and modifies the target machine |
Backing store, in practice
This is the usual answer for "I added a tool or changed a config and want it next time." The project documents creating a backing store to keep your modifications, and it also covers configuring the system with YAML files and running your own scripts at start-up with autorun. A concrete scenario: you regularly recover Windows machines, so you add a notes file, tweak network settings and drop a custom script into autorun. With a backing store on the USB stick, all of that is there the next time you boot it.
For custom software specifically, the documentation also covers installing additional packages with pacman. Those additions are what a backing store preserves; without one, they vanish at shutdown.
When installing to disk makes more sense
If the machine is yours and you want a permanent rescue environment, installing SystemRescue on the disk avoids carrying a stick and keeps changes natively. It is heavier than a backing store and less portable, so reserve it for a fixed workstation or server you administer.
A useful next step
Decide first whether portability matters. If yes, follow the backing-store chapter in the project documentation and test it by rebooting once before relying on it. If no, read the install-to-disk chapter and confirm you have a spare partition. The Quick Start Guide on SystemRescue is the right entry point if this is your first time using the toolkit.
How do I set up PXE network booting to run SystemRescue on multiple machines?
SystemRescue supports PXE network booting as a documented advanced-use option, so you can run it on several machines from a central server instead of preparing a USB stick for each one. The project's own documentation lists "PXE network booting with SystemRescue" as a chapter, and the toolkit is designed to boot without installation, which fits a network-deployment model where each client loads the rescue environment over the LAN.
What the setup involves
- Get the boot files. Download the SystemRescue ISO and extract the kernel and initramfs from it. Read the Quick Start Guide first if this is your first time using the toolkit.
- Serve them over TFTP. Configure a TFTP server (or a combined PXE service) on your LAN and place the kernel and initramfs where clients can fetch them.
- Point clients at the server. Use your DHCP server to hand out the next-server address and boot filename, or configure each machine's firmware to boot from the network. SystemRescue's boot options can be passed as kernel parameters, which is how you control behaviour per boot.
- Test on one machine first. Boot a single client and confirm the environment loads before rolling out to the rest.
The project's documentation covers boot options and network configuration in detail, and the SystemRescue site is the authoritative source for the current PXE chapter and kernel parameters.
Practical considerations
- Bandwidth and scale. Every client pulls the kernel and initramfs over the network. On a small fleet this is trivial; on dozens of machines booting at once, plan for the load on your TFTP host and switch ports.
- State and persistence. A PXE boot is stateless by default. If you need to keep changes between sessions, the documentation describes creating a backing-store to preserve modifications, and autorun scripts to run your own commands at startup.
- Firmware differences. Older BIOS and newer UEFI machines may need different boot-file entries on the DHCP server. Check each model's firmware settings if a client refuses to boot.
- Security. PXE has no built-in authentication. Keep the boot server on a trusted network segment, since anyone who can reach it can boot the rescue environment and access disks.
A concrete scenario
An IT team managing twenty office workstations could keep a PXE server on the internal network. When a machine fails to boot, a technician selects network boot in firmware, loads SystemRescue, and uses the included tools — GParted for partitions, fsarchiver and filesystem tools for data, and network tools to back up files to a server — without hunting for a USB stick.
Decision criterion
Choose PXE when you regularly service many machines on one LAN and want a single, updatable boot image. Stick with a USB stick for one-off repairs or machines on networks you don't control.
User reviews (0)