Linux gets its own Blue Screen of Death, and it seems more helpful than the one on Windows

News
By Mark Tyson
published

It works by parsing emergency log messages to the BSOD.

Windows BSOD
(Image credit: Shutterstock)

Linux users are getting their own blue screen of death (BSOD) as part of an update to the popular system. However, this Linux tribute to the harbinger of woe on Windows has some cool tricks up its sleeve: meaningful error messages and a QR code for users to quickly get more information on the error they encountered.

Version 255 of systemd, a system and service manager for Linux, was released a few hours ago debuting the "systemd-bsod" component. Phononix reports that the stable systemd 255 release has made it just in time for H1 2024 Linux distribution releases. Those on rolling-release Linux distributions will see these BSODs sooner, assuming they encounter an error.

In the systemd 255 release notes, we see that the new "systemd-bsod" service will “show logged error messages full screen if they have a log level of LOG_EMERG log level.” Reportedly, this service came about via an Outreachy 2023 project.  Most OSes have detailed error logging, and the "systemd-bsod" service will attempt to make sense of the data divulged in the logs.

Windows BSOD QR-Codes

Microsoft added a QR-code to BSODs in 2016, but the Windows code always heads to the same root page: http://windows.com/stopcode. At the time of writing, that static page just tells you to restart Windows in Safe Mode, then open Get Help, type in ‘Troubleshoot BSOD error’ and respond to three questions by clicking buttons. If you haven’t updated a driver, added new hardware, or used Windows Update recently you will receive a message that reads: “We are sorry we couldn't fix the problem.” There is no place to input the stopcode listed elsewhere on the blue screen into the Get Help Windows BSOD trouble shooter.

To make things easier for the smartphone totin’ crowd, the Linux BSOD will also display a QR code to quickly jump into helpful solutions on your smart device.

Microsoft added a QR code to BSODs in 2016, but it isn’t context-sensitive to your particular system crash. It remains to be seen whether the Linux version, courtesy of systemd 255, is more fully featured.

The systemd developers have also introduced some other new features in version 255 that are worthy of note. For example, a new "systemd-vmspawn" tool provides the same interfaces for Virtual Machines as systemd-nspawn provides for containers. There is also the new support for TPM2, and disk encryption based upon it via "systemd-pcrlock.” New options and broader support also feature among the numerous updates delivered to systemd 255.

Mark Tyson
Mark Tyson
Freelance News Writer

Mark Tyson is a Freelance News Writer at Tom's Hardware US. He enjoys covering the full breadth of PC tech; from business and semiconductor design to products approaching the edge of reason.

6 Comments Comment from the forums
  • bit_user
    It should be noted that kernel panics are exceedingly rare, in my experience. It's been at least 5 years since the last time I saw one, and I've probably seen only about a half dozen in the past 2 decades.

    That said, I don't run either bleeding edge hardware or kernels. I'm sure people who use "rolling release" distros experience a few more problems, but that's the tradeoff vs. not having to do periodic upgrades.
    Reply
  • mgast
    This is precisely why I moved from an Arch based distro to a Debian based distro (MX Linux). That said I may have to give Mint Debian a try.
    Reply
  • aerkusa
    I've been on a rolling release for 4 or so years now and I don't believe I've ever suffered a kernel panic either. Maybe I'll have to force one just to see this mythical screen.
    Reply
  • Viking2121
    bit_user said:
    It should be noted that kernel panics are exceedingly rare, in my experience. It's been at least 5 years since the last time I saw one, and I've probably seen only about a half dozen in the past 2 decades.

    That said, I don't run either bleeding edge hardware or kernels. I'm sure people who use "rolling release" distros experience a few more problems, but that's the tradeoff vs. not having to do periodic upgrades.

    I have one system that I cannot get it to agree with Linux and it kernel panics at random, some distros more than others, or it just outright freezes, that seems to be way more common than a Kernal panic which isn't helpful either, I've had windows just not agree with some systems either despite it being stable in stress tests, other systems its rare. Some combinations of hardware/software are just more prone to problems. Thats all it comes down to really.
    Reply
  • bit_user
    Viking2121 said:
    I have one system that I cannot get it to agree with Linux and it kernel panics at random, some distros more than others, or it just outright freezes, that seems to be way more common than a Kernal panic which isn't helpful either, I've had windows just not agree with some systems either despite it being stable in stress tests, other systems its rare.
    Perhaps it's a hardware problem. It could be anything from bad RAM to a PSU that might not be under-spec'd or have any problems of its own, but does a poor job of absorbing line noise or transients.

    Viking2121 said:
    Some combinations of hardware/software are just more prone to problems. Thats all it comes down to really.
    There are obviously more issues on certain hardware than others, but unless you're on some fairly bleeding-edge or fairly obscure hardware, it's probably not the software that's at fault.

    I will caveat that by saying that if it has a Nvidia graphics card, you should really be using the proprietary drivers. Nouveau is virtually unmaintained and the new open source driver isn't quite ready for prime time. However, in just about every other case, the open source drivers are preferable.
    Reply
  • wingfinger
    There was one linux system I was using once that thought it was a good idea to throw a bunch of crap, I mean useful debugging data, into something like UEFI variables. This used up a lot of space and caused a shortage. If you saw this was happening, you could use some commands to delete some of those variables, so you had free space again.

    Hopefully, someone has decided that this was not a good idea, to enable for everyone.
    Reply