CERN Renounces RHEL in Favor of Debian | Hacker News Reader

CERN Renounces RHEL in Favor of Debian

infoq.com

36 pointshilux16 comments

Is there a reason almalinux or rockylinux wasn't used? I would except CERN to want first class Selinux support no?

Well, those are functional clones of RHEL/CentOS and unsuitable for CERN as mentioned in the article.

Best I can tell the article only decries the deprecation of x86-64-v2 in RHEL, and says that for that reason Alma is not a viable solution...

Buuuut here is the fix: http://mirror.grid.uchicago.edu/pub/linux/alma/10.2/isos/x86... There's you Alma installer with support for the processors they use, and you can even install EPEL packages built for x86-64-v2 on Alma as well!

So...I agree with the user above, what stops what I must imagine would be a simpler switch from RHEL to Alma instead of Debian?

Sure, but you never know when they might drop stuff like this. With Debian you know it's in the main iso/core/whatever

RHEL is EOLing CPU architectures that CERN still relies on heavily.

> RHEL 9 mandated x86-64-v2, requiring instruction sets like SSE4.2 and POPCNT, while RHEL 10 targets x86-64-v3. CERN's accelerator infrastructure manages more than 2,200 specialised front-end computers and 17,000 embedded devices, including legacy Core 2-era and custom industrial boards, engineered for 10- to 15-year lifecycles aligned with long-term accelerator shutdown windows. Upstream microarchitecture mandates threatened to abruptly obsolete thousands of functional, purpose-built control nodes.

Okay, but Alma rebuilds with x86-64-v2 as a separate ISO, and they even build EPEL packages with support for it: https://almalinux.org/blog/2025-06-26-epel-v2-now-covers-alm...

So the original commenter's questions stands IMO: why is Alma not considered a solution?

Nobody was ever fired for choosing Debian.

CERN wants x86-64-v1 binaries, not v2.

Ah, you're right! Somehow I "autocorrected" that when reading TFA, now it makes more sense.

And they are, in fact, planning to continue to use RHEL and/or AlmaLinux on their desktop and datacenter machines, even going so far as to adapt Fedora's build system to support building Debian packages[1] to leverage their existing RPM infrastructure.

As I understand it (from reading another article[2] posted on HN a few days ago), this change applies purely to the layer of machines that more or less directly interface with their experimental hardware, which has requirements more akin to industrial controllers than workstations or servers when it comes to the expectation of long service life with limited windows for scheduled maintenance.

[1] https://gitlab.cern.ch/linuxsupport/rpms/koji-debian-plugin

[2] https://lwn.net/SubscriberLink/1092512/0772b817c369632b/

This is a dumb and inaccurate headline.

I often think of Debian as a bit moldy from some changes over the years with legacy support (kludges?) in place, where it seemed more like an outdated Arch Linux at some point once it had switched to systemd.

But if I were CERN, I understand the choice of not getting involved with newer forks that aren't novel distributions you can trust to be around for the long haul. I would choose Debian as well in their place.

Debian13 is quite ok and fresh. 12 was trashy. And systemd is purrfect. If you want to edit ridiculous init files, well, go and torture yourself.

Agree, systemd might be the best tooling framework to ever have cropped out of linux

Alma and Rocky both require x86-64-v2 CPU as a minimum (same as RHEL 9). CERN want baseline x86-64-v1 which Debian provides.

week ago story OP;

Previously: https://news.ycombinator.com/item?id=49562262

Reply on news.ycombinator.com