30 thoughts on “Blue Screens Of Death Are Coming To LINUX”
Linux users overreacting to a minor change that will literally not affect 99.997% of them? Whaaaat, omg, that has never happened before! I am in shock.
Regarding the hate of Microsoft Windows BSOD is the fact that the provided "message" never results in an actual fix after several hours if not days of researching what the hell the 0x0 code is and what the "message" is supposed to be about.
Example: if you're playing a video game and it caused an issue with the GPU and it crashes your system to a BSOD, well, you're BSOD message is going to be something about the GPU doing something stupid, when it really was a game error that caused it. You're going to waste hours if not days trying to research the issue and come up with no fix for the problem. This is why often times most people just reboot and it runs just fine!.
To be fair when you get a BSOD (or a similar system error/crash) restarting and seeing if it happens again is a perfectly reasonable approach. For all you know the BSOD had nothing to do with your hardware/software but was instead caused by a cosmic ray flipping a bit in memory since the vast majority of PCs don't have error correcting memory.
"Don't shoot the messenger" was my exact thought when you started talking about it. One thing though, will this BSOD not allow the user to click through and ignore the error if they so choose? Say, if they know through expertise how to solve the issue without rebooting. I don't agree with that at all. User freedom must be preserved, at all costs.
In my experience, and the reason I hate BSOD, is because the messages are usually totally cryptic. Maybe 1 in 10 give useful troubleshooting information but usually just meaningless error codes & memory addresses. Of course, sometimes I can search online for the error code and get more information but not always.
I fail to see the point of BSOD on Linux. We already get good debugging information when something crashes either in the form of useful error messages or good logging…Not to mention solid community support. Seems like QR codes will be a nightmare to maintain and add very little value vs just searching for the error message online.
I guess I've never met anyone stupid enough to think the BSOD itself was the problem. The main complaint I've heard is that it provides very little useful information (to the layman). I'm reasonably confident that the systemD folks will be able to do a better job, and make it actually good.
Calling it a blue screen of death is 100% a troll, though. Nicely done.
The windows blue screen is one of the most useless things in the world. "Hey, something went wrong, here's a useless error code with 100 different issues and 10000 solutions that won't help in your case. Good luck." What I like about Linux is that I can see exactly what's wrong in logs and solve a specific problem.
Yeah even before I got into IT and became a programmer, as a user I was telling people that the bsod isn't an issue of itself but just a helpful message that stuff broke hard. The only reason I actually hate it, is because it's blue and it's objectively the worst color. Apart from my objectively correct opinion on colors tho, it's just annoying and even panicking to get a flash of painful cold fullscreen of blue color. Hope we'll have bsod themes on Linux.
Wait, people blame bluescreen for shit??? I always thought it was just a useful error reporter. Hell, bluescreen is the only reason I found out I had a corrupt memory stick.
Yeah, a square code that I have to install another system on a device I don't have to read is a much more easily accessible way of telling me what's wrong with my computer than just writing it all out on the screen…
How exactly will this interact with verbose boot logging?
I always remove the "quiet" setting from the kernel cmd so i can see everything that's happening during bootup.
Will this BSOD just clear the screen and remove all the usefull information to replace it with some stupid QR-Code which is literally impossible to make sense of without a separate device?!
For me this sounds like it's trying to fix an issue caused by bad configuration practices.
Your view of the windows blue screen of death is *way* too generous. The reason I switched to linux was that my laptop kept crashing with a bsod. I'm not averse to the idea of fixing it myself, so aside from reading the displayed text (basicaly "youw computew had a pwobwem. pwease weboot!") I googled the error code that the bsod so helpfully provides. The gist of it also boiled down to "could be drivers, could be a virus, we don't know, lmao". Assuming the device a total failure I installed Linux. Upon the first reboot it too crashed – but not to a bsod devoid of information. No, it said "could not find inode on device /dev/sda1. Run fsck /dev/sda1 manually to fix". So, as a complete Linux noob I ran the command, and the system successfully restarted!
BSOD is a really usefull feature that microsoft fked up with the latest iterations of windows (from no time at all to read wtf is it saying, to doing the stupid bs of having to take out my phone and scan the QR code), but linux users are against such feature that could be better? wut?
The biggest issue with BSoDs on Windows (at least in the XP era – I've not daily drove a Windows machine since the days of Vista) weren't that they appeared.. but that what little information they provided was cryptic at best (driver filename and internal error name like PAGE_FAULT_IN_NONPAGED_AREA) and just completely useless at worst (nothing more than the STOP code). Doesn't help that, at least in my install of XP back in the day, the default action the machine would perform on a BSoD is an immediate, automatic reboot before you could see it. Woke up to a desktop that had spent half the night boot looping (try to boot -> BSoD during -> immediately reboot -> BSoD….) from a stick of RAM going bad (which took ages to track down… memtest86 with 2-3 passes had no error… from a power cycled boot. Reboot into it via the reset switch and it started throwing errors immediately) and ended up with a corrupted system32 somehow. TRWTF, as always, is Windows.
If the implementations using systemd-bsod can provide something of debugging value, then they've succeeded.
I wish the kernel itself would go further and scribble over the GUI if it panics. It's no good writing any sort of stack trace in panic (possibly too unsafe to write to disk) situations on a VT the user can't see or switch to. If it's broken beyond usability I want to see why if possible, not just a freeze.
Providing info about errors is fine and I support it. As long as it's possible to ignore it and continue using a possibility unsafe system, for those who might for whatever reason need that functionality (eg. if you know one of your drives that should be there on boot is dead) – unless it's a Kernel Panic, of course, for these are fundamentally unrecoverable. If such an option is there (like it was in the old Windows), then I'm all for it. Would've enabled it on my own machine, if not for the fact that I'm on Void Linux (although with Wayland's continued adoption and Void's unwillingness to provide basic support for it in DEs (and I ain't running a WM – I have a life), I may unfortunately end up switching to something else soon – which sucks because I love Void, but the X11 ship is sinking fast and I don't want to be stuck with it, should some apps I use drop support). If it's not – well, SystemD missed the point of Linux itself completely. The whole thing about this OS is putting the user in control of their device, not the other way around. If the user, after being thoroughly informed about the risk, wishes to take it anyway – the machine MUST comply.
I despise this solution. Not having the global (but too little information) poiint of failure is just better. So I want my kernel panics, my regular boot error messages and all the NORMAL stuff – especially bullshit is the… QR code? Wtf is that? I don't wanna install QR reading software on pinephone I want logs and bootloader prompts.
The main issue with BSOD is that it lacks information. For example, if kernel has an assertion issue, a simple BSOD will not tell which file of what line the assertion fail, and thus it will be much harder when someone will post that message/screenshot (with systemd-bsod) to a dev forum.
I generally don't expect Linux users that they don't even can screenshot the error and post it to a dev forum (especially which distro you run). How they heck systemd can make a QR code that will go to a wiki, and how anyone can expect to that wiki? I don't get it.
Also, kernel panic rarely happens, because it has a mechanism called OOPS, which handles drivers errors, and can be configured while building the kernel. I know this will not happened to most users' but I still don't like copy pasting Windows thing on Linux, especially this.
Linux users overreacting to a minor change that will literally not affect 99.997% of them? Whaaaat, omg, that has never happened before! I am in shock.
This sounds super silly to me and not sure why it would be needed compared to a regular error message. Luckily i use void linux now lol.
openrc if you can hear us please save us openrc please get systemd away from us
This is a great move NGL, people crying about it need to touch grass
Regarding the hate of Microsoft Windows BSOD is the fact that the provided "message" never results in an actual fix after several hours if not days of researching what the hell the 0x0 code is and what the "message" is supposed to be about.
Example: if you're playing a video game and it caused an issue with the GPU and it crashes your system to a BSOD, well, you're BSOD message is going to be something about the GPU doing something stupid, when it really was a game error that caused it. You're going to waste hours if not days trying to research the issue and come up with no fix for the problem. This is why often times most people just reboot and it runs just fine!.
"BSOD" shouldn't be the name we use for it on Windows. It should be called the "fuck you" screen, because it gives you so little info on the problem.
the problem is not with BSOD
the problem is the bug that caused it,
BSOD is a big win
Do you think arch would ever provide openrc as an install option?
Black screen with gray text please, because blue background is hard on the eyes.
To be fair when you get a BSOD (or a similar system error/crash) restarting and seeing if it happens again is a perfectly reasonable approach. For all you know the BSOD had nothing to do with your hardware/software but was instead caused by a cosmic ray flipping a bit in memory since the vast majority of PCs don't have error correcting memory.
"Don't shoot the messenger" was my exact thought when you started talking about it.
One thing though, will this BSOD not allow the user to click through and ignore the error if they so choose? Say, if they know through expertise how to solve the issue without rebooting. I don't agree with that at all. User freedom must be preserved, at all costs.
In my experience, and the reason I hate BSOD, is because the messages are usually totally cryptic. Maybe 1 in 10 give useful troubleshooting information but usually just meaningless error codes & memory addresses. Of course, sometimes I can search online for the error code and get more information but not always.
I fail to see the point of BSOD on Linux. We already get good debugging information when something crashes either in the form of useful error messages or good logging…Not to mention solid community support. Seems like QR codes will be a nightmare to maintain and add very little value vs just searching for the error message online.
I guess I've never met anyone stupid enough to think the BSOD itself was the problem. The main complaint I've heard is that it provides very little useful information (to the layman). I'm reasonably confident that the systemD folks will be able to do a better job, and make it actually good.
Calling it a blue screen of death is 100% a troll, though. Nicely done.
The windows blue screen is one of the most useless things in the world. "Hey, something went wrong, here's a useless error code with 100 different issues and 10000 solutions that won't help in your case. Good luck." What I like about Linux is that I can see exactly what's wrong in logs and solve a specific problem.
Yeah even before I got into IT and became a programmer, as a user I was telling people that the bsod isn't an issue of itself but just a helpful message that stuff broke hard. The only reason I actually hate it, is because it's blue and it's objectively the worst color. Apart from my objectively correct opinion on colors tho, it's just annoying and even panicking to get a flash of painful cold fullscreen of blue color. Hope we'll have bsod themes on Linux.
Wait, people blame bluescreen for shit??? I always thought it was just a useful error reporter. Hell, bluescreen is the only reason I found out I had a corrupt memory stick.
"Useful" doesn't always mean "Used"…
Yeah, a square code that I have to install another system on a device I don't have to read is a much more easily accessible way of telling me what's wrong with my computer than just writing it all out on the screen…
You're a god damn moron if you believe that any of this matters. It's open source. Fork it if you don't like it.
Did it really have to be blue? BSOD was supposed to be red…
How exactly will this interact with verbose boot logging?
I always remove the "quiet" setting from the kernel cmd so i can see everything that's happening during bootup.
Will this BSOD just clear the screen and remove all the usefull information to replace it with some stupid QR-Code which is literally impossible to make sense of without a separate device?!
For me this sounds like it's trying to fix an issue caused by bad configuration practices.
This guy is low-key cringe ngl
Your view of the windows blue screen of death is *way* too generous. The reason I switched to linux was that my laptop kept crashing with a bsod. I'm not averse to the idea of fixing it myself, so aside from reading the displayed text (basicaly "youw computew had a pwobwem. pwease weboot!") I googled the error code that the bsod so helpfully provides. The gist of it also boiled down to "could be drivers, could be a virus, we don't know, lmao". Assuming the device a total failure I installed Linux. Upon the first reboot it too crashed – but not to a bsod devoid of information. No, it said "could not find inode on device /dev/sda1. Run fsck /dev/sda1 manually to fix". So, as a complete Linux noob I ran the command, and the system successfully restarted!
There might be arguments about exactly how SystemD does this, but the raging at the idea itself is asinine.
BSOD is a really usefull feature that microsoft fked up with the latest iterations of windows (from no time at all to read wtf is it saying, to doing the stupid bs of having to take out my phone and scan the QR code), but linux users are against such feature that could be better? wut?
The biggest issue with BSoDs on Windows (at least in the XP era – I've not daily drove a Windows machine since the days of Vista) weren't that they appeared.. but that what little information they provided was cryptic at best (driver filename and internal error name like PAGE_FAULT_IN_NONPAGED_AREA) and just completely useless at worst (nothing more than the STOP code). Doesn't help that, at least in my install of XP back in the day, the default action the machine would perform on a BSoD is an immediate, automatic reboot before you could see it. Woke up to a desktop that had spent half the night boot looping (try to boot -> BSoD during -> immediately reboot -> BSoD….) from a stick of RAM going bad (which took ages to track down… memtest86 with 2-3 passes had no error… from a power cycled boot. Reboot into it via the reset switch and it started throwing errors immediately) and ended up with a corrupted system32 somehow. TRWTF, as always, is Windows.
If the implementations using systemd-bsod can provide something of debugging value, then they've succeeded.
I wish the kernel itself would go further and scribble over the GUI if it panics. It's no good writing any sort of stack trace in panic (possibly too unsafe to write to disk) situations on a VT the user can't see or switch to.
If it's broken beyond usability I want to see why if possible, not just a freeze.
Providing info about errors is fine and I support it. As long as it's possible to ignore it and continue using a possibility unsafe system, for those who might for whatever reason need that functionality (eg. if you know one of your drives that should be there on boot is dead) – unless it's a Kernel Panic, of course, for these are fundamentally unrecoverable. If such an option is there (like it was in the old Windows), then I'm all for it. Would've enabled it on my own machine, if not for the fact that I'm on Void Linux (although with Wayland's continued adoption and Void's unwillingness to provide basic support for it in DEs (and I ain't running a WM – I have a life), I may unfortunately end up switching to something else soon – which sucks because I love Void, but the X11 ship is sinking fast and I don't want to be stuck with it, should some apps I use drop support). If it's not – well, SystemD missed the point of Linux itself completely. The whole thing about this OS is putting the user in control of their device, not the other way around. If the user, after being thoroughly informed about the risk, wishes to take it anyway – the machine MUST comply.
I despise this solution. Not having the global (but too little information) poiint of failure is just better. So I want my kernel panics, my regular boot error messages and all the NORMAL stuff – especially bullshit is the… QR code? Wtf is that? I don't wanna install QR reading software on pinephone I want logs and bootloader prompts.
The main issue with BSOD is that it lacks information. For example, if kernel has an assertion issue, a simple BSOD will not tell which file of what line the assertion fail, and thus it will be much harder when someone will post that message/screenshot (with systemd-bsod) to a dev forum.
I generally don't expect Linux users that they don't even can screenshot the error and post it to a dev forum (especially which distro you run). How they heck systemd can make a QR code that will go to a wiki, and how anyone can expect to that wiki? I don't get it.
Also, kernel panic rarely happens, because it has a mechanism called OOPS, which handles drivers errors, and can be configured while building the kernel. I know this will not happened to most users' but I still don't like copy pasting Windows thing on Linux, especially this.