SUSE Announces Its Forking RHEL, To Maintain A RHEL-Compatible Distro

[bug77]

ORACLE says they want to invest $10M to avoid vendor lock-in.

Hold the internet, I'm gonna go look for a snowstorm in July for the next 2 weeks.

No need to look for a snowstorm in July. Oracle is just protecting the platform on which they lay their vendor lock-in. It's a normal reaction.

[pWe00Iri3e7Z9lHOX2Qx]

My thoughts exactly. But I'd have thought SUSE would be better off staying where they are, as the situation might have lead to people moving away from RHEL-based distros.

Timing wise I don't think SUSE was in a great position to benefit from Red Hat's latest communication fiasco. The future is a bit too uncertain with ALP. I don't think people managing enterprise IT budgets are going to be excited about throwing money at SLE today with a ~2 year horizon before ALP arrives with plenty of unknowns at this point.

[Estranged1906]

Sounds like they're giving up on SLE to me

[abufrejoval]

For example the SAP stack is only certified for RHEL, Oracle AAAAND Suse. I wouldn't say that it's a niche or irrelevant.

Actually, numbers are both hard to come by and hard to trust. I guess it's personal perception that somehow SuSE or SLES had fallen from grace, perhaps because they held onto Xen much longer, or perhaps because of Novell.

After starting with Microport on my 80286 I had gone with UnixWare for some years, then with FreeBSD and only went with Linux when it lost its pimples.

It was SuSE mostly because it was domestic and lying around in booshops everywhere, but I didn't use it as a principal system at the time: that was actually Windows by then and the various other *nix variants (including Solaris for x86) were dual-boot options, resting on trayless hot-swap media for a bit of dabbling every now and then.

On the job it was more HP-UX and AIX, even Solaris.

Later, when I became leader of the architecture team, I had my company switch from diverse local variants (SuSE in Germany) to CentOS globally and perhaps that has blindsided my perspective a bit, too.

SuSE used to be the default in Germany, but that home-team advantage suddenly seemed to melt away when they were sold to Novell and in that international context I didn't want to favor any of the European champions, CentOS seemed more palatable, both a neutral and an economical choice.

And actually we ran CentOS on an OpenVZ kernel, not an option with SuSE or many others.

Lately it's been mostly oVirt HCI clusters, first on CentOS7, currently on Oracle Linux 8: that platform's future was already cancelled when CentOS became Stream.

Today I'm glad SuSE survived and hope they can join the post-IBMhat Enterprise Linux Alliance!

[pWe00Iri3e7Z9lHOX2Qx]

They've offered this for at least a year now. It's called Liberty Linux. I still think it's stupid and would rather them spend the $10M either branding and advertisement or on KDE, HDR, and Wayland so they can offer the best desktop experience. The logic is by winning developers and end users over with their superior desktop offering, those users will word of mouth SUSE into becoming the de facto enterprise choice.

What's next? Are they gonna start making forks of every other distribution? Liberty Arch? Liberty Gentoo? Liberty RebeccaBlackOS?

But copying your competitor? That's only going to make people wonder what makes the competitor so special and to probably go with them instead of you.

AFAICT Liberty Linux isn't really a "distro" that you can download / buy and install so the name is a bit confusing. It's more like an umbrella support provider for mixed environments along with management / automation tools. So in your hypothetical example, there wouldn't be a "Liberty Arch Linux". The existing Liberty Linux offering would just provide support and the tooling / automation would be extended to support Arch.

[pWe00Iri3e7Z9lHOX2Qx]

this is the most expensive trolling move in the recent past!

good job Suse!

The $10M USD figure was actually really interesting to me. That's almost nothing in terms of investment these days. The loaded cost of a dev for this is probably north of $200K USD annually. Maybe they can fully absorb the project infrastructure wise with what they already have for SUSE / OpenSUSE / openQA / OBS / etc. But still, $10M goes fast these days.

[abufrejoval]

AFAICT Liberty Linux isn't really a "distro" that you can download / buy and install so the name is a bit confusing. It's more like an umbrella support provider for mixed environments along with management / automation tools. So in your hypothetical example, there wouldn't be a "Liberty Arch Linux". The existing Liberty Linux offering would just provide support and the tooling / automation would be extended to support Arch.

Confusing it certainly is: I've just tried to download it again, and it's the only variant you can't download from SuSE...

So whatever it is (or was) it doesn't seem to vy for a leadership position against Alma or Rocky, while the press at least at the time seemed to believe it was yet another EL clone.
And those included a distro switch script as default add-on, which to me is a minimal requirement to contend in that market.

[pWe00Iri3e7Z9lHOX2Qx]

But why? They already have well established enterprise distribution. Why waste money to fork another enterprise distribution?

The rationale is explained a bit better in the article linked from main one here: https://www.suse.com/c/at-suse-we-make-choice-happen/

The best way to explain this is by the following comparison:

If you are a mobile phone user, you want the ability to switch telco provider while keeping your number, to maximize the value you are consuming.

Equally, as an Enterprise Linux user, you can switch to SUSE while keeping your existing Linux. At SUSE, we are experts at providing enterprise value to users of open source software in a highly competitive way without compromising what is important to customers.

Emphasis mine. So I imagine it goes something like this...

• Our management / automation tooling already supports RHEL
• We hard fork RHEL 9.2
• We promise to support this 9.2+ fork for at least as long as RH was going to support RHEL 9.x
• We provide conversion scripts
• You convert and we support you on this fork, presumably at a lower cost than what RH is charging you today
• You are of course welcome to transition to SLE / ALP over time, at your own pace

This sounds like a reasonable market expansion approach for SUSE.

[acobar]

With all the consolidation happening toward clouds and virtualization on them, I guess, they have data suggesting that consolidation is also happening at used/cilent OS level, and it seems to be around RHEL. Better get prepared to what is coming.

Today is really common to see companies where their servers are almost all from cloud vendors, and I bet there is really a high push by whoever develops for them to pick specific configurations.

[dekernel]

It will act as migration platform for many org. Then they will slowly transfer them to their SEL base with some discounts

Interesting way to look at this move. I really hope it works out for them because I simply love SuSE so hoping this doesn't come back and bite them in the arse.