Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I suspect the lawsuit is created by lawyers, not techies.

Equally reporting on this whole issue seems to be by journalists, not techies. It's been framed (a lot) as a Windows issue not a Crowd Strike issue.

(8.5 million machines were affected, out of 1.4+ billion windows machines [1])

I have one affected customer (10k machines) who assumed I'd suffered like he did, and was surprised when I said we weren't affected. The reporting was consistently that it was a Windows issue, caused by an MS update.

Even this article leans into this narrative...

"as the New York Times put it, “It is more apt to ask what was not affected.” The answer is Linux, Macs, and phones."

Let me add "not to mention 99.4% of windows".

So the journalists don't know what happened, or who was affected, and felt "some computers have a problem" was a weak headline. The lawyers get that narrative and run with it.

And yes, it's easy to squint and claim the "OS should cope with this", but there's realistically limits on what an OS can do once you install a kernel-level driver on the machine. Should we go after Intel for making the chips?

[1] https://www.pcworld.com/article/608447/microsoft-delighted-b...



Reports in the mainstream media were absolutely insane. All the language used points to some kind of unlucky event similar to a bad weather pattern. I knew the general IT knowledge isn't very high but I didn't expect newspapers to report on it like a tornado or an earthquake...


"Let me add "not to mention 99.4% of windows".

...And neither were any of my Windows 7 systems affected.

"And yes, it's easy to squint and claim the "OS should cope with this", but there's realistically limits on what an OS can do once you install a kernel-level driver on the machine."

What are those limits, and why are they limits? Are there solutions? Yes there are. For instance, Microsoft doesn't takes snapshots of the working system state before loading a kernel patch, which on crash it would automatically reload without patch, nor does it employ various other techniques that would solve the problem.

I've discussed these issues in other posts so I won't repeat them here.


It's arguably also a fault of MS: https://news.ycombinator.com/item?id=41096344


> Windows is unsuitable precisely because it can be brought down by third party updates

If I run bullshit on a Linux or MacOS box it can also be unstable and brought to its knees. Or is that poster really trying to argue there's no way you can get a Linux box to lock up?


Key quote from my link:

> Third party vendors are forced into writing unsafe kernel drivers because Microsoft does not provide sufficient user mode APIs.

AFAIK it's different on Linux, and the reliability is higher. Is this not the case?


https://forums.rockylinux.org/t/crowdstrike-freezing-rockyli...

https://access.redhat.com/solutions/7068083

https://lists.debian.org/debian-kernel/2024/04/msg00202.html

You can install buggy kernel modules in Linux as well. I can't even count how many times an apt upgrade/yum update made my system unbootable when using nvidia GPU drivers.

And besides, if you're really wanting that AV system to deeply know about everything the operating system is doing and hook into tons of syscalls, you pretty much can't be running exclusively in usermode. If someone compromised the root of the system you then can't trust the info the kernel is giving your usermode application. eBPF isn't usermode.

And in the end, the poster literally said Windows is unsuitable because third party updates can kill it. That was the key takeaway from their post. Well, third party updates can kill Linux, it can kill MacOS, it can kill darn near everything.


It's an argument. I'm not sure it's a _good_ argument, but hey it's an argument.




Consider applying for YC's Winter 2027 batch! Applications are open till November 2.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: