Enterprise Linux: Are We There Yet?

Simon Crosby writes " Network Computing is running an special report on Linux in the enterprise. It evaluates strengths and weaknesses of Linux useage in the enterprise. It also discusses perceptions, roadblocks, security, clustering and other Linux enterprise issues."

Comments

Lack of Apps

[Jeff Kelly]

Are we there yet? certainly not. Linux has shown in the last few years that it is an alternative, although it still cannot compete in every Aspect with commercial Un*x Systems. (Especially Performance)

The greatest drawback for using Linux in your Enterprise is not the Performance issue but lack of Applications. Many Porting efforts are still beta, (Or do you consider Oracle to be stable on Linux?) or simply not done.

It is still difficult to convince the big software firms to actually consider Linux as an alternative, especially in the Enterprise computing field.

There has still much lobbying to be done.

Jeff

In certain areas...

[alfredw]

I would go so far as to say that Linux is now the choice solution for enterprise web servers - Apache is all its glory, etc.

Doubly so, given IIS's press lately.

marketing

[kraada]

my thought is that linux will eventually pick up, but it's going to take a while. Why? Linux works great and people will eventually realize that, but there really isn't much marketing as far as I can tell. IBM with it's Peace Love and Linux campaign has probably done a lot, but compared to MS . . .
*shrug*
Once linux's PR is going well, poor MS will probably be in for the fight of it's life . . .

We may not be enterprise, but...

[omega9]

It has been, and will continue to be right for us:

I am one of two people who run the MIS office at a small technical college. We support everything that plugs into the wall for ~500 staff, faculty, and students. We're not huge, but we keep busy :) We run a Windows 2000 Adv. Server domain that seems to run well for the most part but acts... haunted, as if it's just scared to run stable all the time.

Recently we decided to replace an old outdated MS-Proxy server (don't yell, I know) with a Squid server. Not because squid is more stable, but it was damn near impossible to monitor MS-Proxy cheaply and effectively. We farmed around for sollutions but they were all over $1,500 for anything decent. Any of you familiar with Squid can finish this story yourself. For the others, now we can monitor at any depth all we want for $free (and of course it's more stable).

Last night I sat that same server up with MRTG to monitor all of our switches and the network interfaces on all the servers in our farm. Oh, and it's also a syslog server, watching over the the network printers and the server farm itself ( we have Event Viewer in Win2k setup to dump to a CSV file and then push it to syslog). It's also a planned backup www/ftp server and what-ever-else-is-needed server.

Proxy, MRTG, better monitoring then Event Viewer, syslog, www/ftp: all on one box, at zero software cost. Since it's Linux I know I can trust it to keep up. And I'm not saying that because I'm wearing my shebang(#!) hat and my tux hoodie, I'm talking from experience. You can't beat it.

That's why Linux is here for us now, and it's because of those qualities that Linux will continue to grow with us in the future.

The definition of 'Enterprise'

[Binary Tree]

Somebody please define it, in the context of software. What makes linux any less "Enterprise-ready" than NT, AIX, Solaris, etc?

Does it even have a real definition, or is it just nonsense like the term "supercomputer"?

Re:Are we there? Will anyone ever be there?

[fanatic]

(Some is also legal... if you run into a snafu with kernel 2.6.1, who can you sue??).

You sound like you've got a good view of the issue, but this sentence cries for rebuttal. When, oh WHEN, will pople stop parroting this nonsense? Any CIO that uses this as an argument against OpenSource/Free software is a moron. I challenge anyone, anywhere, to give evidence that anyone has ever collected a single penny from suing a mass-market software maker for shoddy code. If MS didn't lose their shirt over putrid crap like win3.x or win9x, with it's dll-hell and semi-annnual re-install schedule, how can anyone get sued?

Re:"who do you sue?"

[fanatic]

a better question is: if things go wrong with widget x, what are my options to get it fixed? with closed s/w, the only option is the vendor you got it from (and really, knowing that, do you want to sue them?). with free software you can use your vendor, another vendor, your own staff, or private contractors

Someone, please, mod this up some more.

Re:MS Word

[joeytsai]

This is true, but unfortunately, this is very, very, very hard. Speaking as someone who's worked in the filters area of a major company's word processor (think what used to be Word's competition back in the day) filters are horribly complex and making a quality filter is just grueling... and we're not even talking about a "100%" one.

In fact, even Microsoft's filters aren't very good. Naturally, most people don't even notice it because .doc has become the de facto standard.

And this is where the situation is bad: if filters aren't that great in commercial word processors, with full time paid programmers, they will be long coming in Linux. Let's be honest, people working on projects for fun will probably have the itch to do something much more interesting and noticable.

But you're right, .doc is the standard on the vast majority of computers today, so have great import filters is needed. However, I think it would be a better idea for the focus of the different Office suites to be a common file format. All the groups - Gnome, KDE, OpenOffice, should really decide on a XML based format and work together on the .doc import.

How are we going to compete with the Windows world, which has one common file format (albeit terrible) when our own different office suites each have their own file format (which will naturally mean even more filters)?

Office Apps != "The Corporate Enterprise"

[brassrat77]

I think MS Word interoperability is perhaps the single most important barrier limiting companies from changing to Linux. Other Office products such as MS Excel and MS Powerpoint are also important.

Desktop office applications are a noticable but small part of "the Enterprise" and NOT the main point of the original article.

"Enterprise" usually refers to the core applications running in the corporate data center. Inventory, payroll, order processing. Applications where downtime costs $$/minute. Applications where "No application"=="No business".

Linux is making gains in these areas. The adoption rate appears slow because

1. It is slow. "Bet the company" decisions are always slow. Implementation is slow. Anyone remember how long it took Windows NT to break into corporate data centers? (Many would argue it still isn't ready)
2. Companies don't always consider what is a "mission critical" application. Areas where Linux excels - web, mail, dns, and many other RFC-based services, for example - may not be viewed as "critical". At least until the boss wants to know why the corporate web site is down (nmida) or the email system is hosed (badtrans). Then we get something like the Giga Group recommendation to use anything but IIS.
3. Companies see this as a competitive advantage and do not want to discuss it. The big NYC financial firms are a good example.
4. Consulting firms need more linux experience. Many enterprise customers rely on the IT consulting arms of the big system integrators and consulting firms. If these outfits push something other than Linux, something other than Linux gets proposed. Do they get incentives from MS, Sun, IBM, HP, CPQ...? Maybe. Anyone pushing Linux like that? Not yet.
5. CIOs don't always know what's running. I've come across repeated examples where the top managers swear "Linux isn't allowed" but there are stealth, pilot, and production deployments all over the shop. The file sharing and print system runs Samba on Linux and "just works" (and isn't considered "enterprise" until a key document is needed).

Penetration of Linux could still be better, of course. We need better support from enterprise management and backup systems. We need more "mind share". This article helps.

Desktops remain a problem. Out of sight, out of mind. Windows is in everyone's face every day.

Still Does Not Answer My Question

[4of12]

This is all fine and good - using Linux for servers is a great business decision. No licensing hassles, stays up like a champ and keeps on performing. End of story. Let's move on.

But what about:

Doing system administration for large LANs of Linux desktops?

Over the years we've been running RISC workstations that are becoming increasingly expensive from a hardware standpoint relative to what can be got in the x86 world.

We'd like to take advantage of the price performance advantage in hardware as well as the increasing maturity of Linux desktop end user applications (which are getting real close now). It seems like a lot more applications are available for Linux desktop than many of the traditional commercial Unices.

The problem is that everyone I know that runs Linux runs their workstation or laptop as their own cowboy system administrator. They typically don't worry about integrating dozens or hundreds of these things together in such a way that a small support staff can manage them effectively.

You know the kinds of systems.

• Haphazard applications installed whereever they felt like
• distro installed out of the box without enough applications
• no patches applied for necessary security updates
• strange hardware hanging off moldy interfaces,
• never thinking about whether it might be nice to use something like NIS (but I know that's not good enough) or automount,
• never providing regular user file backup,
• deciding whether to put apps on a central server or each desktop, (pros and cons either way)
• how to handle upgrades,
• etc.

So what I want to know is:

Has anyone done this?

How did it work? What should we look out for? What is the advantages and disadvantages? Good tools? Web sites?

Almost There

[uslinux.net]

There are still a few things lacking in Linux distributions to make it fully "enterprise-ready" (I hate that term). For anyone who has truly spent time with other Unices, the following is obvious - Better Package Management! The ability to test out packages without "commiting" them so, if something breaks, you can immediately roll back to the prior state. Yes, I know you can uninstall and reinstall old packages, but it's NOT the same. Use HP-UX for a while and you'll understand - you can install, remove, commit, rollback, and test packages. In a production environment, it is critical that a newly installed patch or program doesn't break existing systems!

As far as the kernel goes, I think Linux is there. I DON'T think Linux is necessarily ready to compete with NT or 2000 (though I give it 18 more months), since it is still lacking quite a few easy to use admin tools (think of the NT print manager or DHCP admin and you'll understand what I mean), but it is coming along.