Alma and Rocky Linux release 8.5 builds, Rocky catches up with secure boot AlmaLinux and Rocky Linux, both of which provide community builds of Red Hat Enterprise Linux (RHEL), have released builds matching RHEL 8.5, with Rocky's work catching up with Alma by being signed for secure boot. Would-be CentOS replacements AlmaLinux and Rocky Linux track RHEL closely, and differ from CentOS Stream in that …
Still feel leery that one or both of these will not survive more than 1-2 years, or at least will not survive in the community/free model they espouse currently. After checking out the Rocky Linux AMA on Reddit, and hearing about challenges that the CentOS community ran into over the years with keeping enough skilled volunteers doing work to survive, I can't help but see similar fates ahead here. Remember that CentOS was propped up since 2014 after becoming part of Red Hat, where Red Hat paid those engineers market competitive salary's and benefits to do CentOS work.
My questions are:
1. If the CentOS original developer community was overwhelmed with the amount of work, while not getting paid, and working other jobs to support themselves to have time for CentOS work, how will that be different if Rocky and Alma are trying to build a community of volunteer developers to build, maintain and respond ?
2. As time goes on, do you think there would be a split of community members supporting Rocky Linux for free on their own time, and others supporting Rocky Linux but getting paid for it via Greg Kurtzer's other for-profit other company or other company's possibly ? And similarly with Alma Linux having unpaid volunteers, and those at parent Cloud Linux getting paid to do similar work ?
I am thinking that the ones doing it for free will begin to feel a bit chump-ish. And then the community support or people creating the builds for no pay could die away, and what we would be left with is a maybe-cheaper version of RHEL from both companies, and the logic of "we have to pay people to do timely support and updates" ?
I kinda don't get it, since CentOS did not work out as a viable long term plan, and as my grandma used to say "there ain't no free lunch", and people hoping that "all will be free and high quality forever" seem out of touch to me.
It may be useful to know that AlmaLinux's migration script can migrate from Rocky in addition to CentOS and the other RHEL clones, and similarly Rocky's migration script can migrate from Alma in addition to the others.
I don't see that as squabbling over each other's user base but instead a very sensible insurance policy for the users should one or the other go to the wall.
I am mainly questioning what to recommend for business customers that have been using CentOS, so that they don't face a similar migration decision in 1-2 years. And I can't see how Rocky or Alma stay solid and viable over time, due to the volunteer requirements to keep things solid, and at no cost. In addition ISV's have to decide if they want to support one, or both of them, in addition to RHEL, CentOS for awhile longer, Ubuntu, and maybe Suse. Which most ISV's don't want to keep adding test and certification and support platforms to their matrix from. my discussions with them.
Color me skeptical still, and trying to figure out what to do. Most of the customers I advise are on CentOS 7, so they have some more time to play things out. The few on CentOS 8 I have suggested migrate to RHEL for now, since I could not confidently recommend an alternative at this early stage of things.
It is all fascinating and interesting though....
I think that at the beginning a bit of healthy competition is probably healthy. Whether there will prove to space for both of them longer term, only time will tell.
AlmaLinux is coming from the creators of CloudLinux which was already based on RHEL (or was it CentOS) so they may not need huge additional resources to make AlmaLinux work. According to Wiki Rocky is being funded by Amazon Web Services, Google Cloud and Microsoft Azure among others.
One reason why one or both of these maybe viable when CentOS proved not be is that IBM/RH pulling the plug on CentOS may have scared some of the MegaCorps running data centres large enough to have their own weather system, full of CentOS system who have been eating at the free lunch buffet table for years to put their hands in their pockets and toss some small change in the direction of keeping these distros afloat this time. The alternative would be either paying RH or rolling their own.
We're just going ahead with a split RHEL 8 (where we want outside support) and [CentOS|Alma|Rocky] 8 (where we don't).
In that model, AlmaLinux and Rocky are more like alternative repos than different distributions. We 'll switch our internal mirror (repo server) between them as needed. Individual Linux hosts just pick up whatever is on today's menu.
Just feeling that for places where they can shift or change the OS with limited impact it is ok to experiment on things like Alma and Rocky, but when you have an entire application stack and security policy, based around the OS, and you want to run it on-premise, and in clouds, and you want to ensure various hardware and software are compatible - CentOS had become pretty safe on all those checkmarks. It was only a challenge if one needed support or up to the minute security patches.
I am leery that anything can replace that, and am thinking CentOS was a "once in my lifetime phenomena" combination of things - brewed in the early years of commercial open source and intertwined with Red Hat being a rare company that can "sell free software" successfully.
Will all be part of the fascinating history of open source software...
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
