Incorrect: the backdoored version was originally discovered by a Debian sid user on their system, and it presumably worked. On arch it's questionable since they don't link sshd with liblzma (although some say some kind of a cross-contamination may be possible via a patch used to support some systemd thingy, and systemd uses liblzma). Also, probably the rolling opensuse, and mb Ubuntu. Also nixos-unstalbe, but it doesn't pass the argv[0] requirements and also doesn't link liblzma. Also, fedora.
Yes, but Arch, though it had the compromised package, it appears the package didn't actually compromise Arch because of how both Arch and the attack were set up.
I thought Arch was the only rolling distro that doesn't have the backdoor. Its sshd is not linked with liblzma, and even if it were, they compile xz directly from git so they wouldn't have gotten the backdoor anyway.
The extent of the exploit is still being analyzed so I would update and keep your eye on the news. If you don't need your computer you could always power down.
In the case of Arch the backdoor also wasn't inserted into liblzma at all, because at build time there was a check to see if it's being built on a deb or rpm based system, and only inserts it in those two cases.
i think it’s a matter of perspective. if i’m deploying some containers or servers on a system that has well defined dependencies then i think Debian wins in a stability argument.
for me, i’m installing a bunch of experimental or bleeding edge stuff that is hard to manage in even a non LTS Debian system. i don’t need my CUDA drivers to be battle tested, and i don’t want to add a bunch of sketchy links to APT because i want to install a nightly version of neovim with my package manager. Arch makes that stuff simple, reliable, and stable, at least in comparison.
"Stable" doesn't mean "doesn't crash", it means "low frequency of changes". Debian only makes changing updates every few years, and you can wait a few more years before even taking those changes without losing security support while Arch makes changing updates pretty much every time a package you have installed does.
In no way is Arch more stable than Debian (other than maybe Debian Unstable/Sid, but even then it's likely a bit of a wash)
Just Arch users being delusional. Every recent thread that had Arch mentioned in the comments has some variation of "Arch is the most stable distro" or "Stable distros have more issues than Arch".
It literally does though. Stable doesn't mean bug free. It means unchanging. That's what the term "stable distro" actually means. That the software isn't being updated except for security patches. When people say stable distro, that is what they are trying to communicate. That means the software will be old. That's what stable actually means.
In my experience they're the same from a reliability standpoint. Stuff on Arch will break for no reason after an update. Stuff on Debian will break for no reason after an update. It's just as difficult to solve reliability problems on both.
Because Debian isn't a rolling release you will often run into issues where a bug got fixed in a future version of whatever program it is but not the one that's available in the repository. Try using yt-dlp on any stable Debian installation and it won't work for example.
Arch isn't without its issues. Half of the good stuff is on the AUR, and fuck the AUR. Stuff only installs without issues half the time. Good luck installing stuff that needs like 13+ other AUR packages as dependencies because non of that shit can be installed automatically. On other distros,all that stuff can be installed automatically and easily with a single command.
You can get yay for an AUR package manager, but it's generally not recommended because it means blindly trusting the build scripts for community packages that have no real oversight. You're typically advised to check the build script for every AUR package you install.
Stuff on Debian will break for no reason after an update
I have never had this happen on Debian servers and I've been using it for around 20 years. The only time I broke a Debian system was my fault - I tried to upgrade an old server from Debian 10 to 12. It's only supported to upgrade one version at a time. Had to restore from backup and upgrade to Debian 11 first, then to 12.
Arch has already updated XZ by relying on the source code repository itself instead of the tarballs that did have the manipulations in them.
It's not ideal since we still rely on a potentially *otherwise* compromised piece of code still but it's a quick and effective workaround without massive technical trouble for the issue at hand.
From what I read it was one of the contributors. Looks like they have been contributing for some time too before trying to scooch in this back door. Long con.
Basically, one of the contributors that had been contributing for quite some time (and was therefore partly trusted), commited a somewhat hidden backdoor. I doubt it had any effect (as it was discovered now before being pushed to any stable distro and the exploit itself didnt work on Arch) bjt we'll have to wait for the effect to be analyzed.
There are no known reports of those versions being incorporated into any production releases for major Linux distributions, but both Red Hat and Debian reported that recently published beta releases used at least one of the backdoored versions [...] A stable release of Arch Linux is also affected. That distribution, however, isn't used in production systems.
I think the confusion comes from the meaning of stable. In software there are two relevant meanings:
Unchanging, or changing the least possible amount.
Not crashing / requiring intervention to keep running.
Debian, for example, focuses on #1, with the assumption that #2 will follow. And it generally does, until you have to update and the changes are truly massive and the upgrade is brittle, or you have to run software with newer requirements and your hacks to get it working are brittle.
Arch, for example, instead focuses on the second definition, by attempting to ensure that every change, while frequent, is small, with a handful of notable exceptions.
Honestly, both strategies work well. I've had debian systems running for 15 years and Arch systems running for 12+ years (and that limitation is really only due to the system I run Arch on, rather than their update strategy.
It really depends on the user's needs and maintenance frequency.
Not gonna lie, this whole debacle made want to switch to NixOS.
Immediately rolling back to an uncompromised version was my first thought.
That and the fact that each application is isolated from each other right? Should hopefully help in cases like this
It's definitely common, but zstd is gaining on it since in a lot of cases it can produce similarly-sized compressed files but it's quicker to decompress them. There's still some cases where xz is better than zstd, but not very many.
(and please, absolutely don't run above as root. Just don't.)
I carefully answered to retain any root owned files and my backups, despite knowing the backdoor wasn't included in the culprit package. This system has now "un-trusted" status, meaning I'll clean re-install the OS, once the full analysis of the backdoor payload is available.
Edit: I also booted the "untrusted" system without physical access to the web, no gui, and installed the fixed package transferred to it locally. (that system is also going to be dd if=/dev/zero'd)