Search
  • Michael R Sheehan

CentOS is Not a Tech Problem; it is a Business Problem – A Big One


The Demise of CentOS Will Cost Money, Guaranteed


If your business runs on CentOS - your CIO, Fractional CIO, IT Team, or IT Resources are going to have to deal with a thorny issue.


CentOS was the most utilized Linux server operating system on the Internet.


Take a moment to truly understand the breathtaking scope of CentOS in the world economy, U.S. business enterprises, almost every industry you can name; governments, financial, healthcare, military. Hundreds of millions of servers, providing exabytes of data through an operating system that was stable, free, and reliable.


Some very recognizable names include Amazon, Amazon AWS, Whole Foods, Google, Twitter, Facebook, Wikipedia, GoDaddy, Rackspace, Disney, NASA, JPL, Dow Chemical, Johnson & Johnson, Bayer, Novartis, NYSE, London Stock Exchange, US Department of Defense.


But It Is Not Just the Big Guys


From a size of all the customers that are using CentOS, 46% are small (<50 employees), 31% are medium-sized and 22% are large (>1000 employees). From a revenue of all the customers that are using CentOS, a majority (65%) are small (<$50M), 18% are large (>$1000M) and 9% are medium-sized. According to Enlyft.


And that is just the companies that were, or are, using CentOS to run their servers and infrastructure. But CentOS is EVERYWHERE.

"This year (2018) 71.8 percent of respondents ticked "Linux" to the choose-all-that-apply question, what operating system(s) do you use for your IoT devices?" according to the IoT Developer Survey.


The Internet of Things (IoT)


Much has changed since back in 2013 when the Internet of Things (IoT) emerged as the next shiny tech bauble. The challenge in understanding IoT is not only that the technology is constantly evolving, but that the definition keeps expanding from an original focus on machine-to-machine (M2M) applications that communicate without human intervention.


Linux-driven open source and commercial single board computers or modules sit at the heart of the IoT phenomenon. They are usually found in the form of gateways or hubs that aggregate sensor data from typically wirelessly enabled, sensor-equipped endpoints.

  • Drones

  • Wearables

  • Automotive

  • Aviation

  • Home Automation Hubs

  • Surveillance

Those extremely affected are pharmaceutical, biotech and life-sciences organizations, whose M2M connections are in the trillions. Open source communication, scientific data processing, and cloud hosting have all embraced CentOS as a godsend. One less thing to be a hassle.


All of these users, including development teams, need stability. Now CentOS latest release, instead of a stable life cycle and support system that had nine years left to go, CentOS 8 will expire at the end of this calendar year. There will not be a CentOS 9 or later.


What’s So Special About CentOS?


Linux is free.

So is water, but that does not stop companies from trying to differentiate and sell you.. water.


Linux has many, many variants, and a couple of different distribution stems.


The significant distribution here is Red Hat, founded in 1994, who were able to profit from the free resource of Linux by offering service & support. Red Hat did well in stabilizing and popularizing Linux in general, and their kernel in particular. Their customers enjoyed tech support, and a vast community.


Then In 2004 Red Hat made two major changes. First provided the community of open-source code enthusiasts a variant called Fedora, which was the community’s to use as a sandbox. Coding enthusiasts began building all sorts of interesting things. Users created many of the improvements, upgrades, and improvement elements that made their way into future iterations of the Red Hat Linux OS. Which led to the second change, Red Hat decided to add proprietary and trademarked coding to its Linux kernel and package it for resale as Red Hat Enterprise Linux. The improvements from Fedora became defined as an upstream Red Hat Enterprise Linux also known RHEL.


Immediately thereafter, Gregory Kurtzer took the RHEL operating system and copied it with goal of removing the proprietary and trademarked code. This is known as a software fork or just fork. Community Enterprise Operating System or CentOS was born.


CentOS benefitted from both its roots in RHEL and being intrinsically tied to RHEL, upgrading on a schedule just months behind its closest relative. This is defined as a downstream operating system.


CentOS was always updating with production released code from Red Hat. Therefore it was certified a “stable” version of RHEL, it was free, and it offered the computing world a perfect alternative to licensing or SaaS. Essentially a clone of RHEL with all of the benefits and none of the costs, CentOS solved a lot of problems.


As the web browser became more and more of a universal interface, usage of CentOS exploded. Developers relied on CentOS, built around it and its stable 10-year life cycle (one of the longest in the industry), they relied on steady improvements and release cycle of Red Hat, and CentOS became the Linux OS of choice for much of America.


Then, in 2014, Red Hat joined the board of CentOS. The community that ran CentOS was having challenges with ownership of URLs, Trademarks, and source code locations. Red Hat officially sponsoring and joining the CentOS board brought needed stability to the operating system’s development life-cycle and release. It also gave Red Hat voting power to control CentOS. As this Ars Technica article puts it “... an acquisition in all but name – Red Hat now both funded and controlled CentOS…


This was not a problem until IBM purchased Red Hat. You could see where this was going.


The Vicissitude


Just months after its acquisition by IBM, in September 2019, Red Hat announced CentOS Stream. The idea was to use CentOS Stream as a midstream between the upstream development in Fedora and downstream development in Red Hat Enterprise Linux (RHEL). If Fedora was the beginning and RHEL was the profit production release, then CentOS Stream was the middle.


Then in December of 2020, Red Hat changed everything. CentOS Project Board released a statement that CentOS Project will “be shifting focus from CentOS Linux, the rebuild of Red Hat Enterprise Linux (RHEL), to CentOS Stream.” In that same statement, CentOS Project Board announced that CentOS 8, the latest release of CentOS, End of Life (EOL) has been moved from 2029 to 2021.


In an interview with TheRegister, Brian Exelbierd, Community Business Owner, RHEL Product Management at Red Hat, told them “the CentOS board doesn’t get to decide what Red Hat engineering teams do.” Also mentioned in that interview “CentOS Linux is ending because Red Hat simply refused to invest in it.” Fairly definitive on what is happening next for CentOS. To repeat – there will not be a CentOS 9 or later.


Anyone involved in road mapping CentOS 8 life-cycle with nine years left now have ten months. After that no longer will corrections and improvements flow down from RHEL into CentOS.The work stream will be Fedora into CentOS Stream, into RHEL. CentOS is simply cut from that workflow.


Already existing life-cycles of older CentOS versions will continue, for now, with the original End of Life dates. CentOS 7 is expected to receive maintenance updates through June 2024, outliving CentOS 8.


The Sky is Not Falling


Yes, this IT issue is going to cost American businesses money, there is no way around that.


Organizations as of today have three obvious options:


Do Nothing

You can continue to use CentOS for as long as it works. Kick the can down the road. Someday you will have to decide which of the other two options you are going to pay for.


But you will pay.


As with all choices, there are consequences. By choosing to do nothing and continuing with what you have, you will be pushing your organization and infrastructure into unsupported and unsecure territory. Not the best position to be in.


Doing nothing is not the same as ‘wait and see.’ Doing nothing simply moves the date you choose to make the inevitable choice. If you do nothing for too long the choice may be made for you with little input or control.


According to a CloudLinux survey, most CentOS users, 60.5%, are waiting for a CentOS fork to be released. Of the rest, 16.7% are looking to Debian Linux, 12.4% are considering Ubuntu, while 10.4% are thinking about openSUSE. It appears the majority of organizations are taking a wait and see approach before deciding between the next two options.


Pay Red Hat

You can absolutely sign on board with Red Hat and get their service and support for the essential OS. Depending on your size and scope, this requires careful financial consideration, but is easily implemented.


Red Hat has announced free production use of RHEL for small workloads of 16 licenses/servers.


It is interesting to point out that this announcement came a month and half after Red Hat announced the end of the CentOS. You only need to read headlines and reddit posts that the CentOS community is not happy with Red Hat’s decision.


Go With One of the Other Options of Linux

CloudLinux has pledged a million dollars a year in development money for their own CentOS/RHEL offshoot, called AlmaLinux. They have already started publishing migration articles and released a beta operating system


Gregory Kurtzer, co-founder of the CentOS Linux distribution, has founded a new startup company called Ctrl IQ, which will serve in part as a sponsoring company for the upcoming Rocky Linux distribution. Ars Technica reports "Kurtzer co-founded CentOS Linux in 2004 with mentor Rocky McGaugh, and it operated independently for 10 years until being acquired by Red Hat in 2014. When Red Hat killed off CentOS Linux in a highly controversial December 2020 announcement, Kurtzer immediately announced his intention to recreate CentOS with a new distribution named after his deceased mentor.


It should be noted that currently Amazon has thrown its considerable support behind Rocky Linux.


Choose to move your organization to another Linux distribution altogether such as SUSE Linux Enterprise Server or Ubuntu OpenStack Enterprise.


Any migration to a new operating system will have consequences.


For larger organizations, it will not be simple to migrate off CentOS/RHEL based infrastructures.


Your enterprise environment with have workflow and process controls - not to mention third party utilities - built into it for administration. Support update packages and release tools differ from Linux distribution to distribution and in most cases will not play well together.


So What’s Going to Happen?


“History doesn't repeat itself, but it often rhymes” – Mark Twain


Remember way back in 2009 when Oracle bought Sun Microsystems? Remember MySQL?


Once MySQL development became questionable with Oracle ownership, American business did what it always does - pivoted to the next thing and kept going. Same thing will happen here, business will decide between the three options above and act accordingly.


Neither CentOS, as an operating system, nor any of its progeny, will ever be the monolithic giant it was in fall of 2020. Once fragmented, it won’t ever grow as tall again. Businesses will pay, in the long term or the short term, and end up doing what is best individually.


Whatever your conversion method or eventual platform, you will need guidance. Do not go it alone. The choice you make in a post CentOS world will have ripple effects on your business technology, the skills of your operational IT staff and in some cases, development of your products being sold. Discuss this issue with your IT resource, your peers, even your competitors, then make your choice.


Based in Dedham, MA, ConaLogix was founded in 2018 as a fractional CIO and advisory resource for the pharmaceutical, life science, and biotech industries. We provide C-level Information Technology services on a virtual basis, assisting with architecture, integration, and testing. Using a unique, customized approach, the ConaLogix team collaborates with entrepreneurs and scientific core teams to support their vision, while guiding the most efficient development model that benefits data management and core requirements


What Are You Doing With Your Data?


79 views0 comments