Red Hat Linux Support To End
Orbital Sander writes "Received a missive this morning from the Red Hat Network, stating that they will discontinue maintenance on Red Hat Linux 7.x and 8.0 by the end of 2003, and on Red Hat 9.0 by the end of April, 2004. And, more ominously: 'Red Hat does not plan to release another product in the Red Hat Linux line.' [The full text of the email is on Newsforge.] Kind of the end of an era, and the new king has already been appointed: Red Hat Linux is dead! Long live Red Hat Enterprise Linux! Looks like they realized that only their support contract-based version of the product was making them any money." Readers also note that Red Hat is pointing users to the free Fedora Project.
so long, and thanks for all the RPMs.
Are you secure enough in your masculinity to run 'man touch'?
From where I'm standing this looks like a very silly step on Redhat's behalf.
I have two Redhat boxes at the moment, one running 7.1 which handles mail and DNS for me a half a dozen friends/family, the other running 9.0 which is purely a remote backup server (rsync copies data to it daily).
I use Redhat because despite the fact that I installed 7.1 a couple of years ago I pay my $60 a year so that I can run "up2date" once a day to keep my security patches up to date. I pay my $60 for both systems.
I also buy a copy of Redhat every 18 months or so.
Now that they have decided to stop updating 7, 8 and 9 they are forcing me to migrate both boxes. I don't have time to scan the web looking for security updates for hundreds of packages, so I need an update service. Hell, I only installed the 9.0 box 4 months ago and come next April updates stop !
So it looks like they are forcing me to either move to Redhat Enterprise to get security updates from them. It looks like I would have to stump up two lots of $379 just to get a two copies of Enterprise and 12 months of update for my two boxes.
I obviously don't want to pay that much...
So I guess I'm going to have to migrate to Debian or something instead ?
The end result for Redhat, no more income from me.
I use Red Hat 9 at home. Because of this, when time came to roll out some Linux servers at work and my boss asked me which we should use, I told him "Red Hat Enterprise" (we wanted support and had the money to pay for it).
I suspect that for a reasonably significant portion of their market, Red Hat Linux (and cooresponding useful items like RHN) is the primary reason that their customers buy Enterprise. I hope they've considered this...
Every year during my review, I just pray the words "slashdot.org" aren't mentioned.
This info has been around for a long time. Red Hat Fedora Core 1 was due to be released today, but they found an issue so it's delayed, as you can see from the Fedora schedule. You can read the mailing list post about it here.
This is a bad situation for those of us using RedHat Linux, but there *is* hope.
Check out my eclectic infosec blog at InfoSecPotpou
It means Red Hat isn't going to sell a product in the Red Hat Linux line.
...
It doesn't say there won't be a distribution in the tradition of Red Hat Linux. In fact, Fedora Core 1 is about to be released
The server line only is so successful because of the branding of the desktop line. If they drop one, they'll lose the other. Not to mention that it's Almost to the point that corps will be willing to pay for it! That's great, drop the OS just as it's about to become functional!
The main problem I had when I received this is they seem to be really focused on the "Enterprise" aspect of this. I am a happy subscriber to the update service with a handful of servers. However, none of these boxes are really "servers" in the heavy duty use sense. We use them as firewalls, and one as a light duty PHP/mysql/web server for doing bug tracking, design documents, etc for the developers.
Under the old scheme, I was able to purchase the low end version and run it as a light duty web server. Now, looking at the product mix, it looks like they are taking the Microsoft 'your workstation isn't a web server' approach to stratification.
Sig under construction since 1998.
But you need to offer a service that someone wants in order to make money. I think people would pay for linux, it's a great OS, but when its perfectly legal to just download it and install it for free why would you pay for it? Only if the incentive for purchasing it was good enough. There's been plenty of companies that have tried to make a profit selling linux, but only a few have come out ok. I know everybody is going to bitch about the spirit of free software and all that crap, but the people at Red hat have families to feed too. Sometimes I wish linux was cheap not free. $50 for an enterprise class system is a damn good deal.
BUT They still have and fund the Fedora Project. This is essentially Red Hat linux. It's just no longer commercially supported. Just like debian.
Red Hat will have to continue releasing any GPL'ed code in the same way they always have. You may not get any proprietary software, but I can't think of anything that was, in base Red Hat.
I'm less concerned with the "no new Red Hat" than with "You've got two months to upgrade". Many vendors only support what RH supports, so vendors may no longer support their products on the free system, and that's a big headache for SA's.
Someone asked if I had patched against MSBlast; I said yes, I installed Linux.
Anyone with an elementary school understanding of arithmetic and a lick of common sense can tell you that Red Hat's business model was unsustainable.
A free product, free downloads, free support?
Enterprise linux support? Sure, until it's profitable enough that Big Blue decides to take it from 'em.
Big Blue is the only company around poised to profit from Linux. And we all tip our hats and give them our full support. Hip hip hooray.
Does noone see that the open source community is nothing more than a source of free labour to IBM?
They'll milk Red Hat for free code, and when the work is completed to their satisfaction, they will have the might to succeed where SCO fails - "owning" Linux.
Why do people think IBM is a "good" company? Their track record makes MSFT look like a care bear convention.
I don't need no instructions to know how to rock!!!!
Open source software doesn't feed the family so what do all those out of work developers do? It seems to me that OSS is like a virus that eventually consumes its host, thus ending its own life.
This is a serious question from one who seeks to be educated.
Oh yeah, I already know that I am an idiot and most likely a facist, capitalist, bozo, insertyourlabelhere so save those type of comments for your high school classmates and please seek to address the question.
New Coke.
Probably not their best move to date.
This promises to be interesting. I like RedHat, but mainly because of inertia. I've been running it since 6.2 and haven't been sufficently motivated to change. As a result, when asked what distro to run for professional applications, I say "RedHat" due mainly to farmiliarity.
Microsoft has been rumored to almost encourage "piracy" of their office suite because it leads to adoption by paying customers. RedHat is obviously a stepping stone to RHEL. Without providing a "personal" version, RedHat will be able to devote much more energy to large dollar corporate customers, but the lack of grassroots support may offset the increase.
Seems to me that the product formerly known as Redhat Linux, is now called Fedora. What's the big deal?
Buy Slack distros. I do.
If you don't like Slackware, there are many other distros out there ...
I see a lot of people posting "time to learn the debian install." Perhaps not (even thought its not hard folks) since anaconda has been ported you might soon see Debian install ISOs with a familiar face. I think Debian + Apt + Anaconda destroys redhat as a desktop distro, as the only problem I had with Debian usability wise was the install, keeping updated and secure is as easy as a cron job. Forget a good day in redmond, I think its a good day for Linux not to be tied by ignorant people to the Red Hat name. Then again, I can't wait for the FUD C|Net, Dvorak et. al spew out.
Are you secure enough in your masculinity to run 'man touch'?
My company has over a dozen Red Hat servers, about $900 a year in RHN seats. That's $900 a year Red Hat's getting just for providing us updates, no support.
We're migrating slowly to Debian since this latest Red Hat policy change was announced.
This article pretty much sums up what I am facing.
I've had enough abrasive sigs. Kittens are cute and fuzzy.
What I worry most about is how this will impact corporate perception of "free" software. Even if RedHat decides to back down from this policy, the MicroSoft marketing drones will almost certainly use this as an example of: "Look how crazy those open source nuts are! You never can count of the product to be around long term." Obviously this would be the pot calling the kettle black given MS's record of forced upgrades, but a little hypocrisy seldom gets in the way of an MBA on a rant. :)
I don't understand that. What's wrong with SUSE? SUSE is easily the equal of RedHat. (I'm talking about 8.2 here - I've not tried SUSE 9.0 yet).
Reality is defined by the maddest person in the room
I'd played around a lot with the Fedora Core beta (Severn) over the weekend, and wanted to describe my experience a bit for those thinking of going that route. Purely anecdotal, your mileage my vary, and all that stuff.
I initially installed over an existing RH9 install, and also tried an install on a fresh partition. The install process was very similar and it upgraded my existing packages nicely, and did a good job of preserving configuration.
Fedora also has a couple channels on redhat.com for up2date, they work a lot like the one from RH9, but with newer versions of the software. Initially I was subscribed to the Rawhide channel, but after updating up2date itself, it changed to a Fedora Core channel that offered the same stuff. Four of the packages (the desktop backgrounds, indexhtml, and some http configuration package) did not have the right GPG signature, which causes up2date to prompt you (annoying during a very long download that should be able to complete unattended), and can also make up2date hang when it goes to install those packages.
On a positive note, Fedora can recognize my Broadcom ethernet on its own now, with RH9 I had to download and install a separate driver.
Red Hat Graphical Boot (rhgb) is pretty hit or miss, I had it working briefly but it broke again. Looked pretty good while it was working, but was hard to keep working. Also didn't appear to have much in the way of man pages.
The system would sometimes slow way down when booting as it got to probing modules and/or detecting new hardware. I got errors about it trying to install the floppy.o module (floppyless system), and sometimes lots of stuff scrolling by about other block-major devices not being found.
The Linuxant Driverloader program I need to use my WiFi card installs under the 2.4.22.2088 kernel, but after doing up2date and getting the latest (2.4.22.2115, iirc) it would not install. Even under 2088 it gave me problems I had not encountered when running it on a RH9 system that had been updated to the same kernel.
When doing an update install, it adds a new entry to your existing bootloader, as would be expected. When doing a fresh install, it seems to only let you use GRUB, which could be an annoyance to those who prefer LILO. Of course you could change it after the fact.
To sum it all up, Fedora Core is for the most part quite slick and I really liked that it has more current versions of the packages than RH9, which has to play it safe for the corporate world. However, I experienced enough frustrations to have doubts as to whether Fedora Core is really as ready as it needs to be to take over from Red Hat 9.
"- Open source healthcare"
/.?
Is that going to be anything like the Open Source legal advice you get here on
geek#1: I have this rash on my ass and I can't find anything in the man, wiki or google about it. Anyone have an idea what this is and what I can do about it?
geek#2: IANAD but...
I think the biggest problem with Fedora Core is that it doesn't associate itself by name either to Red Hat or to Linux, the two biggest branding assets in the Linux world. D'oh!
You say "Linux" or "Red Hat" to the electronics store geeks and they finally know what you are talking about these days. You can tell your boss that you want to run "Red Hat Linux" and he'll consider it.
Now you have to go to the electronics store and answer the "What kind of computer do you have?" question with "I use Fedora Core." Will your boss consider letting you use "Fedora Core 1" even if you promise him that it's really "Red Hat Linux 10" in disguise?
Why not "Red Hat Fedora 10?"
Why not "Fedora Linux 10?"
Why instead the relatively obscure "Fedora Core 1?"
And it's a very awkward phrase... Think of the authors of "For Dummies" books who will how have to say "in Fedora Core, XYZ" over and over in their books instead of just "in Linux, XYZ" so as not to confuse the reader!
And will readers that set out to buy books about Linux even figure out that they now want the book about "Fedora Core?"
Similarly, most of the people that I know who have considered toying with Linux know only about Red Hat Linux. When they finally get a free afternoon and try to locate it, will they make the connection and figure out to download Fedora Core 1 over their broadband connection, or will newbies be downloading Red Hat Linux 9 for the next four years because it's the highest numbered Red Hat Linux they can find?
Seems like a dumb marketing move, as far as I'm concerned.
STOP . AMERICA . NOW
The worst part of EOLing the RedHat line is that there isn't a real migration path from RedHat to RedHat Enterprise. Basically, the migration path is 1) back everything up, 2) install RedHat Enterprise, 3) restore user data such as home directories, databases, mail configuration, etc. 4) spend the next week getting the server to work as it did before you installed RedHat Enterprise.
If you're trying to migrate a critical installation that can't be down for long periods of time, I guess you're SOL.
I work for a large WebHosting Company. I'm not due to start work for a few more hours, but I can already imagine some of the things that must be happening.
We have thousands of servers, hundreds of them are RedHat Linux. Our Flagship Systems Management product runs on RedHat Linux and FreeBSD. Our model has been very effective and efficient so far, because RedHat Linux had known reliability and cost factors. With Cost about to skyrocket, and a limited migration opportunity timeframe, we're screwed. Many other organizations who chose RedHat Linux for similar reasons and deployed it in similar numbers are screwed as well.
IMHO this is a bad move for RedHat only because of the no advance notice. Had they said this 6 months ago, everyone would be in a better position to deal with it.
My company can not, and does not, just go around upgrading all the servers. We do them when the box fails, customer has problems, or is hacked. This is the only time when the customer feels that a change is necessary. No one has the time to migrate en masse.
RedHat does want our money, I can assure you. Though we haven't paid them much, many of our customers have. Plus, we help give them Name Recognition. Customers come to us for our excellence of service (we are actually that good), and if they choose Linux they get RedHat. They learn more about RedHat and coupled with our quality, they will probably continue on in life very happy with the idea of using RedHat Linux.
Now we have to start figuring out what to do.
Thanks RedHat. Your loyalty to your customers is crap.
Next time, how about just two weeks for the End of Life announcement.
Q: What is the errata policy for The Fedora Project?
A: Security updates, bugfix updates, and new feature updates will all be available, through Red Hat and third parties. Updates may be staged (first made available for public qualification, then later for general consumption) when appropriate. In drastic cases, we may remove a package from The Fedora Project if we judge that a necessary security update is too problematic/disruptive to the larger goals of the project. Availability of updates should not be misconstrued as support for anything other than continued development and innovation of the code base.
Red Hat will not be providing an SLA (Service Level Agreement) for resolution times for updates for The Fedora Project. Security updates will take priority. For packages maintained by external parties, Red Hat may respond to security holes by deprecating packages if the external maintainers do not provide updates in a reasonable time. Users who want support, or maintenance according to an SLA, may purchase the appropriate Red Hat Enterprise Linux product for their use.
1. "Everything" will never be written. To think that OSS will have written everything, and there's no commercially viable programs left is silly.
2. In-house developments and/or adaptations of OSS work requires programmers. In fact, most programmers today are busy doing in-house things.
3. There's always some things for which there is more money than programmer interest, which simply wouldn't be written unless those with money paid for it. Think uncool, boring, tedious, repetitive programming with hardly any value to the general public.
Besides, there's nothing fundamentally wrong or unique about the process destroying the market. Think e.g. a company that has specialized in automating manufacturing - replacing humans with robots. Once they're "done", they've obsoleted themselves, since their services won't be needed anymore.
Except that for them too, the job is never done. All the time new products go from prototype stage (typically with some or a lot of manual labor) into full-automated production, creating new jobs. Same with programming. This program or that has been "done", but there'll be other programs, other software.
Maybe you think the PC and Linux is like the "final" step. In my opinion it is only the beginning, as more and more embedded devices (everything from cell phones to dish washers to PVRs) are becoming "mini-computers", almost without exception commercial and proprietary (at some level, like OS X over BSD and Tivo over Linux). And all of those will need developers...
Kjella
Live today, because you never know what tomorrow brings
I've just read the posts at +3 and it seems like everyone thinks this is a negative, bad thing -- it's not at all :-)
RedHat have found that a free software project cannot be developed in a close way -- it is too expensive amoung other things. So they have opened up development to the community.
If you just follow some of the mail on the fedora lists you will find that the opening up of the project has led to loads of cool stuff starting to happen, the fedora legacy project to support old versions, people offering to do i18n stuff, people working on a PPC version, support for apt and yum -- none of this would have happened without out the dev being opened up.
Also why is it called Fedora? -- well one reason is so that anyone can duplicate CDs and sell it! Before people doing cheap CDs had to remove the Redhat trademark stuff, now you don't need to :-)
Check out MKDoc a mod_perl CMS
No, it's not. If it costs them $2 to get every $1 from you, they don't want your $1.
What's foolish is some slashdotters' simplistic ideas about business.
Of course that's not to say that this move is going to help them. The side effect of making Red Hat less visible may well cost them more than they save, but if they can't find a profitable business model, they'll disappear anyway.
"Those who have never entered upon scientific pursuits know not a tithe of the poetry by which they are surrounded."
I have a handfule of servers and found it very convienant to pay $60/year per machine to have a centralized place (rhn) to track updates & perform installations.
This wasn't profitable?!?
Fedora's rapid-update cycle ruins it for me, keeping machines on software/releases that are "patchable", without an upgrade, will simply take to much effort
I think you guys don't get it
I think this makes a lot of sence.
1. Technological advances made in Fedora will make it into Redhat Enterprise Linux.
2. RedHat developers will work on Fedora. (Maybe not as many as before)
3. Non-Red Hat Developers can now change RedHat for the better. If you don't like certain things in Fedora you can now change it.
I think RedHat is saying...
We want to concentrate our work on creating the most
- stable
- secure
Linux OS.
I think this is good. Finally there will be a Linux version that you can trust on an enterprise system. I'll bet IBM will jump into bed with this one.
Fedora may suck. But, it doesn't seem that different from the original RedHat.
Redhat just isn't going to spend effort to make it
Robust
Secure
Reliable
Stable
RedHat 6, 7, 8 weren't very stable or reliable in my opinion. And I'll bet the Fedora community could create some sort of update server as well.
I might still migrate away from RedHat. We will have to see what happens. Its all perception... This name change might hurt there image.
Just accept that it is difficult, hence marketing. Bob Young(now gone from RH) said it best. It's all about branding. This will seriously hurt the brand and slow any new blood from jumping on board.
They could have done the same thing structurally and still called it Red Hat Linux. But now people will rightly say, "So why did they change the name?"
Expect to see an attempt at back-pedaling in two years, but it will be to late.
Who will be the next distro king? Who will get all that dirt cheap cross branding for the services their company offers...
Novel theory: Modern Man evolved from psychopath
Though todays announcement shouldn't come as a surprise to anyone who's followed Red Hat over the last year (support discontinuance was announced long ago, Fedora was announced more recently), I think it was a very poor move.
Yes, I do understand producing their "Red Hat Linux" product was expensive, and hurt their bottom line. They should have never split their product in two to begin with. Maintaining both RHL and Enterprise Linux was too much of a burden on the company. It reeks of bad management, much like the Mozilla project does (They are trying to develop no less than three different browsers at the moment, possibly more depending on how you count--and Netscape just cut them lose, so they're severely understaffed... you'd think they'd make consolidation efforts--but this is another tirade).
What they should have done is modularize their base product, and sell add-ons. They retain all of their users, all of their mind share, only have to develop one product, AND it can act as a stepping stone into your Enterprise-level services. Hell! They even had the infrastructure to do a single core product all laid out with Red Hat Network. Sell an Enterprise Web Server channel add-on to Red Hat Linux 10 for Enterprise-level prices, and so on. It would have been beautiful. Really.
It would have also provided their Enterprise customers with ten-times the amount of testing of the core OS. This is not to be underestimated. Much as Linus renames a kernel from e.g. 2.5.79 to 2.6.0-test1 when he wants (free!) wider testing, Red Hat now has a user base one-tenth the size to "test" their releases on. And problems that aren't caught in relase QA (many just can't be) will now HAVE to affect (high-)paying customers. There's no free users to take 90% of the falls.
Red Hat produced the de facto Linux distribution in the United States AND they were in the black. There was nothing to stop them. They provided a free, high quality alternative OS. People were switching to Linux, and switching to Red Hat. It was working. But apparently not fast enough for them.
Windows users have no highly visible, high quality alternative now. (No, it's NOT necessary to chime in with your favorite distribution.) What's good for Linux was good for Red Hat, and this is unquestionably bad for Linux, medium-term, at least.
Fedora does NO ONE any good. It's pseudo-managed by Red Hat, but with no guarantees, no support, no Red Hat Network, no Enterprise add-ons, and regular Joe-Schmoe developers fucking it up (cf. Debian). And the mix of open development and corporate bureaucracy, neither with any vision, is sure to pull and tug at it in no general direction, making it nothing more than a poor Debian clone. I wonder how long until Red Hat cut's it lose completely.
It's a sad day for Red Hat. Up until they split their product line last year, I was considering investing in the company. They had a real handle on the market. Now, they have nothing to drive themselves into becoming a big player. They'll remain a small service-oriented company. If they remain at all. (They kind of remind me of BSDi now. Probably not an association they would like.)
And it's a sad day for Linux. But I have faith the (huge) void will be filled. Will Debian step up? Someone new? It should be interesting, at least.
[Wow. That turned out to be longer than I'd expected. If I wasn't hungover I'd actually invest a little more time and proofread it. Hope it's been an interesting read, if anyone made it this far. Hey, e-mail me if you did! Tell me if you agree, or if I'm crazy, or both. Or just say hi! I'm bored. No one sends letters these days. The Internet's become so impersonal. But that's a whole nother tirade.]
This is, without a doubt, the most horribly incorrect, misleading article I've ever seen.
Real story (a Fedora project member might have additions, but this is pretty close to what's happening). Red Hat realized that Debian was doing something right -- big set of packaged software, auto-updates on 'em, etc. Red Hat was trying to set up their own Debian-style setup called Red Hat Linux Project (RHLP). At some point, Red Hat picked up on the fact that most Red Hat users already like and use Fedora. So they talked to the Fedora folks, and combined RHLP and Fedora. Basically, it means that Red Hat pays for Fedora hosting and distributes Fedora (major value add) *with* their own software packages.
What's the end-user result? From a RH user standpoint, it's something like Powertools being readded to RH plus a lot more. A lot of Debian-style goodness being made available to the masses. There's a much larger package set, so less needing to use checkinstall to automatically produce halfassed RPMs from tarballs. You get a *good* set of download-and-install tools (Fedora uses apt and yum, unlike RH's piss-poor up2date...I've been griping about this on Slashdot for ages). You don't need to add Fedora's apt or yum package to your distro to actually use the large set of well-packaged packages.
This is the *best* thing that's happened for RH users for a long time, and we someone, confused or malicious, posting a "RH is dead" story? What the heck? Is this guy a Mac OS X nut, or just completely and utterly confused?
Clearly, RH should have done a press release, but it was damned irresponsible of Slashdot editors not to add a followup comment, given how significant this is to the Slashdot community. It's like a story claiming "Debian is being acquired by Microsoft" when someone packages WINE for it, or something equally ridiculous.
May we never see th