Which Red Hat Should Be Worn in the Enterprise?

weatherbug asks: "I've recently been appointed as a member of a team to help determine the direction our organization is headed with Red Hat Linux. Currently we're using multiple versions from Red Hat 6.x through Advance Server 2.1. However, now that Red Hat has effectively separated their distributions into a 'consumer' (Red Hat 8,9, etc) and 'enterprise' (Red Hat Adv. Server 2.x, etc), we aren't sure which version we want to adopt. A Red Hat salesman recently told us that the 'consumer' version of Red Hat was mostly for hackers and hobbyists who weren't concerned about stability and wanted the most up-to-date software, while the 'enterprise' version would be more stable and have a five-year product lifetime. As a long time Linux system administrator, I feel that this is a sales tactic and that there really is no compelling reason for us to ever use the 'enterprise' version. After all, it is Linux and it is open source, and we have enough in-house talent to not need Red Hat support. Why would we ever need or care about a five-year product lifetime? Am I wrong, and if so, could you set us straight? We'd be interested to know what other large organizations have decided to do."

Comments

IMHO, you answered your own question

[aborchers]

When you said

As a long time Linux system administrator, I feel that this is a sales tactic and that there really is no compelling reason for us to ever use the 'enterprise' version.

Re:IMHO, you answered your own question

[johnnyb]

It depends on how you answer the following questions:

* Do you want to recompile each package every time you want to update it, or do you want to do rpm -i?

* Do you want to backport source patches to your current version, and then install it, or do you want to do rpm -i?

* Do you want to have to watch every mailing list for possible security problems on your software, or do you just want to look in the errata section of the RedHat web site?

I think there's essentially five options:

* Continually reinstall your servers to the latest RedHat

* Buy Advanced Server

* Form a community group dedicated to keeping up with older versions of RH - making the above changes as a group

* Use a distro that already has a community group (i.e. - Debian)

* Do it all yourself

Or you could just leave your machine unpatched :)

Why bother at all?

[TCM]

If you really have enough in-house talent to not need Red Hat support why bother with Red Hat on a commercial level at all? Just download one of their ISOs (that is possible, right?) - or any other distribution for that matter - and do it all yourself. Correct me if I'm wrong but the number one reason to actually pay for a Linux distribution is the support that comes with it, isn't it?

Are you writing custom applications?

[cgenman]

If you are, you may need support for many years for that OS version 9.x. There can be libraries that your application relies upon, but those older-version libraries might not be present in newer versions of the software that contain exploits you would want to patch, or features you might like to build around.

Food for thought.

And if you don't need Red Hat's service plan... why not just run Debian -> Stable?

Upgrade every year?

[orev]

Think about a mission critical system that needs to run 24x7. Every time you have to apply a patch or upgrade the system, that's downtime you can't afford.

"Enterprise" servers are one's that just work and you don't have to mess with them. That is contrary to what most sysadmins like to do with systems - that is, mess with them constantly.

In 5 years...

[NathanE]

I think one thing to keep in mind is what will your tech department look like in 5 years. Shoot, 5 years ago who would have guessed things would be like they are now? Say your staff is halved in 5 years for whatever reason. Will not having official support matter at that time? I'm not trying to advocate buying Advanced Server, but you should at least keep in mind that crazy things can happen over the course of 5 years.

To some, the extra money is well worth the insurance you get.

For what it's worth

[npietraniec]

For what my opinion it worth... We've got about 200 workstations (a decent enough size network) and we've got several RH servers... we standardize on every 3rd release (6.2, 7.2, and now 9) and don't have any problems. We've got redhat network subscriptions for updates and everything is rock solid. I see no need for "enterprise editions." Upgrading the servers every few years before end of life isn't that horrible for us... And there's usually compelling reasons like journaled file systems and new versions of ssh that justify it.

But you need to evaluate your own needs obviously.

It depends

[darthtuttle]

It depends on how much you rely on RedHat after you install the product, and how much the company wants to continue to do that.

First remember to think in terms of the company. While you and your fellow admins might be uber-gurus you might not be with the company forever. Will they find other slashdot reading uber-gurus to replace you, or will they be left with less capable people?

Then consider what you do on your own. Do you install RPMs from RedHat, or do you "use the source"? Do you update your own kernel? What do you do if there's a security flaw or bug in a software package? Do you use the source or the RPM.

RedHat offers an attractive model for companies who don't want to depend on having "Bob the admin" around and would rather depend on the idea that "RedHat" will be around (the former usually isn't there as long as is around.)

Everyone company has a different culture and answer, those are some of the questions to consider.

It's not entirely a sales tactic.

[Rimbo]

Not every organization, large or otherwise, has the in-house talent to do their own open-source maintenance and support. Maybe they have most of their machines running Windows, maybe not.

Beyond that, a lot of experienced tech executives, having been burned by a lack of support in the past, are not going to chance it without a service contract like the one Enterprise offers.

The arguments for and against are like the arguments for and against buying insurance, because the support contract is a form of insurance. You will never convince me that the full coverage I pay for on my vehicle isn't worth it, because at the moment my car was stolen and totalled, I received more money back than I'd ever paid the insurance company. On the other hand, you'll never convince my girlfriend -- who drives an '83 Accord -- that anything other than the minimum liability insurance the law requires is necessary.

We're both right, because our situations are different.

This is a management question

[lawpoop]

I think from an administration standpoint, you're right on. But you might want to look at it from a management perspective.

What makes your boss feel more secure? Is your boss the kind to totally trust you and your judgement, or do they like to see some 'backup'?

Also, would you like to be totally on your own, or would you like to be able to say "Know what? I'm sick of this problem!" and call up Red Hat support? This could be helpful in shifting blame away from yourself.

Re:benefits

[photon317]

You can get both from standard source on kernel.org as well. I think what the original poster is *really* asking is "What features exist in RH AS 2.1 that are truly unique and worth the money, as opposed to stuff any good linux hacker could've thrown together on his own?". Redhat would do well to answer this.

Re:What My Organization Did:

[stand]

Seriously, if you don't need Redhat support and don't want their Enterprise features, why would you use Redhat on a server? *BSD or Debian would be more reliable, and Gentoo, LFS, or Debian would be much more customizable.

While I agree with this sentiment mostly, you have to realize that to the people that are paid to make these decisions (The Boss, the CIO, whatever) customizable == bad, at least as far as operating system decisions are concerned.

It's impossible to know if you'll always have the expertise to maintain all your wonderful customizations and since, if the operating system can't be made to work, everything fails, it's extremely important that your os configuration be very transparent.

Of course there are sacrifices in power, flexibility and so forth that you make when you choose this route, but to the CIO, ease of maintanence (not to mention replacability of maintenance staff) tends to trump those considerations.

Redhat realizes this and I think this is the source of their success.

Re:benefits

[Thomas A. Anderson]

I think you answered your own question here. Red Hat, more than likely, does not market the enterprise edition to "good linux hackers". Rather, it's for companies that don't have a large *nix I.T. staff, want to save moeny over traditional unix, but expect the support options of traditional unix.

DEBIAN (was: Are you writing custom applications?)

[pergamon]

I thought RedHat was the greatest thing in the world until I tried Debian. I now use Debian 'stable' on servers, 'testing' or 'unstable' on workstations/etc.

It is hard to beat having security patches backported for keeping a system stable. (The other main reason I switched to Debian is that its the only distro that will install run on all the different hardware I use like PA-RISC, Alpha, Sparc{32,64}, and MIPS without jumping through any hoops).

Before anyone jumps on me with a "this other distro is even better", let me clarify that I'm posting this only to say that I think there's a better option than RedHat. In particular, other great distros like Slack and Gentoo that don't have binary package management systems (for better or worse) aren't really comparable to RedHat. Mandrake, from the few days I've used it, just seemed like a flashier and even more bloated version of RedHat.

The only downside I've found to using Debian over RedHat (or the other distros that are based on RedHat) is that some commercial apps are geared towards RedHat and only release RPMs. In particular, Compaq's Linux support software/drivers are almost exclusively in RPM format. Now 'alien' does indeed convert them to installable .debs for me, but hand tweaking is usually necessary.

And yes, it is much easier to use 'apt-get' than dealing with the RHN to get 'up2date' working.

Re:I'm more worried about...

[Natalie's Hot Grits]

"Um, this has *got* to be a troll. First off, any company that doesn't chase and fix bugs should (and will) go out of business."

This is where you don't understand the differences between their "Enterprise" and "standard" editions...

First of all, microsoft doesn't chase and fix bugs, and they are not out of business. In fact, they are the most profiting company in this half of the world (probably the whole world).

Secondly, That is what RedHat means about a "5 year product life" It means that the "Enterprise" edition of their software will be supported for five years, as opposed to RedHat Linux 9.0's support which will last maybe 1.5 years if your lucky.

The point is that if you are a hobbiest, you will want the latest and greatest version of linux. And you will be forced to upgrade to the latest and greatest version if you want support (Read: patches and updates to the software). If you want support (Patches and updates) for more than a typical hobbiest needs, then you need to go with the "Enterprise" version, which will be officially supported by redhat with updates and security patches for at minimum of 5 years.

If you don't need the telephone support and just need updates and patches, I suggest bypassing redhat's services altogether and going with Debian Linux, which has simlar long term support networks in place by default.

The bottom line is that if you go with the "enterprise" version, RedHat will train and maintain a technical support staff that is capeable of troubleshooting your version of Enterprise Linux for five years. They will also release security updates to your version for five years. If you don't go enterprise, no matter what kind of support services you need, your version of Redhat will be defunct (read: no more security updates on unsupported versions) in probably less than a year. This is not good for enterprise servers that don't need to be upgraded to the latest and buggiest software every 9 months.