Practical Technology

for practical people.

July 11, 2001
by sjvn01
0 comments

A rogue’s gallery of denial of service attacks

Tuesday, May 22, started out as just another day at CERT Coordination Center at Carnegie Mellon University in Pittsburgh. By day’s end, CERT, widely regarded as the Fort Knox of computer security, would be knocked off the net by a distributed denial of service (DDoS) attack.

In 2001, even the crème de la crème of network security is vulnerable. If it can happen to CERT, it can happen to you.

You can help prevent DDoS assaults across the Internet and lower your vulnerability to attacks. But if someone really wants to put your business under with a DDoS attack, they will. Microsoft, Yahoo, and Exodus have all fallen to DDoS attacks within the last 12 months; you or your customers could be next.

A rogue’s gallery of denial of service attacks. More>

April 23, 2001
by sjvn01
0 comments

Gotta Make the In-Laws Happy: Caldera and SCO tie the knot

CALDERA AND SCO WILL finally exchange wedding vows next month, and SCO integrators are eager to attend the ceremony.

During the May 4 event, Caldera’s pending purchase of UnixWare, OpenServer and SCO Professional Services is expected to receive shareholder approval.

That’s good news for SCO’s partners. Caldera couldn’t give much support or information to SCO’s resellers as the $23 million deal wound its way through the legal process.

When the deal gained SEC approval last month, it freed Caldera to share some general business plans with SCO’s partners. Once the May 4 shareholder vote is completed, Caldera will likely crank up the volume on its partnering strategy for SCO integrators.

Caldera hopes to give allies and customers the best of both worlds: Linux’s flexibility and low cost at the departmentallevel, and Unix’s rock-solid reliability at the high end of the market.

Still, Caldera will face plenty of challenges. SCO’s customer base has eroded over the years and dramatically declined between 1999 and last year (SP, April 16, p. 30, www.smartpartnermag.com/issues). It’s not “clear that Caldera can get that share back,” says Dan Kusnetzky VP at International Data Corp.

Analysts note that Sun Solaris continues to dominate the high-end market, with Windows 2000, Windows NT and Red Hat Linux enjoying strong success on Web servers and departmental servers.

Caldera CEO Ransom Love’s strategy, previewed at CeBit in Germany last month, is fairly simple. The company is embedding Linux features into Unix, and vice versa. At the same time, Caldera continues to enhance and refine OpenLinux, UnixWare (to be known as OpenUnix) and perhaps even OpenServer. Version 8 of OpenUnix went into beta this month.

Even before the Caldera deal, SCO was bolstering UnixWate to support the Linux Kernel personality (LKP). Going forward, that means OpenUnix and OpenLinux will have the same GNU tools and libraries, which stick like glue to the proposed Linux Standards Base’s specifications. As a result, developers will be able to compile and run Linux applications on top of UnixWare, which could make UnixWare more appealing to business customers and open-source developers.

Caldera will promote OpenLinux for departmental servers and Web servers, while pushing OpenUnix for high-end servers and enterprise systems.

The company hopes to avoid the type of marketing and development setbacks that plagued UnixWare when Novell owned the operating system in the mid-1990s.

At the time, Novell positioned NetWare as a file-and-print server, and UnixWare as an application server. But the UnixWare and NetWare development teams had a poor working relationship, and plans to meld the two operating systems ultimately stalled. Novell ended up selling UnixWare to SCO, where it enjoyed some initial success before losing momentum in recent years.

Meanwhile, questions continue to surround OpenServer, which was SCO’s flagship operating system until the company acquired UnixWare. SCO tried to force OpenServer customers to embrace UnixWare, but many customers scoffed at the idea.

Some SCO resellers hope that Caldera enhances OpenServer with LKP and other Linux-related features. Without those capabilities, resellers say the end of OpenServer could be near.

Software vendors apparently agree. Many developers, particularly database companies, refuse to port their latest applications to OpenServer because it’s considered to be a dead-end platform. “Caldera talked the talk, but we have to see if they will walk the walk,” says a Midwest SCO reseller.

On the corporate side, some of SCO’s established hardware partners applaud Caldera’s strategy. Gary Campbell, CTO of Enterprise Servers at Compaq, says the merger “will strengthen the relationship” between his company and Caldera.

Compaq preloads OpenLinux eServer 2.3.1 on Compaq ProLiant ML330, DL320 and DL360 servers.

Some of SCO’s oldest and best customers have already announced that they’ll be sticking with Caldera.

BMW, one of SCO’s largest customers worldwide, has a lot of confidence in Caldera International, says the car maker’s European director. BMW is buying the newest release of SCO OpenServer 5 licenses to upgrade existing motor testing applications to the new generation of tester systems, according to the BMW director.

Caldera insists that the marriage of Unix and Linux will be a success, but newlyweds tend to be idealistic. Reality usually doesn’t set in until after the honeymoon.

Big Blue Is Red Hot on Unix Trail

Supporting Linux and AIX alike, IBM’s new AIX 5L operating system and p660 and p620 servers are part of the company’s ongoing push to take over the Unix market lead through a combination of direct sales and partnerships.

IBM VP Mike Kerr thinks that IBM’s just released midrange SMP servers, bundled with AIX 5L, will play mainly in existing AIX environments where IT staffers want to experiment with emerging Linux apps on the side.

Other AIX servers have run Linux before. However, the new pSeries servers add SOI (silicon on insulator technology), adopted from IBM’s higher-end zSeries servers, for faster chip performance.

IBM expects to release TPC-C benchmarks this week, showing better results on a six-way p660 system than on an eight-way Sun UltraSparc 3.

On the scalability side, AIX 5L contains elements of Project Monterey, an earlier joint venture between IBM and Santa Cruz Operation (SCO).

To support 5L, IBM has added several hundred new direct salespeople. The company will attack the enterprise market directly, and the small to midrange market through resellers and integrators. ASPs will also be a target for the p660 rack-mount system.

Darren Shallcross, VP of business development for Big Blue partner Solution Technology Inc., says his midmarket corporate customers will try out Linux e-mail, Lotus Notes and firewall systems.

Caldera will also be entitled to sell AIX 5L after its SCO acquisition goes through. As insiders see it, though, more of Caldera’s energies will go to selling its own UnixWare 7, also containing Monterey elements.

First published in Sm@rt Partner

April 20, 2001
by sjvn01
0 comments

Are Trade-Shows Worth Your Time?

I’ve just spent three weeks speaking at as many trade shows. I did my act at a small DTR Business Systems reseller show in Vegas; a middling Seybold show in Boston, with 1,500 guests; and the 70,000-attendee Spring Comdex in Chicago. Guess which show had, by far, the happiest and most satisfied attendees?

If you guessed the DTR show, with about 80 resellers and integrators, you;d be right. Oh, it wasn’t all sweetness and light:

The DTR people, who base their businesses largely on SCO Open Server, were desperate for answers about the Caldera/SCO merger, since approved by the SEC. But at least they got some answers, and a chance to talk one-on-one with their peers as well as with bigwigs from their distributors and vendors.

Fast-forward to Spring Comdex. Key3Media claimed there were 70,000 people there. I didn’t see them. The show floor was tiny; and the jokes that the Waste Expo trade show for the trash business, right across the hall, was more interesting weren’t very funny, because they were true.Off the show floor, things were worse. Many conference speakers simply didn’t show up, leaving their audiences in the lurch.

But, wait, there’s more. Spring Comdex’s ASP Summit was a joke. The forlorn free-lunch ballroom had only two tables occupied. While my ASP (Application Service Provider) panel mates at Seybold argued that the ASP model was coming into its own, you sure couldn’t tell it from Spring Comdex.

It wasn’t just there, either. SP senior editor Fred Aun tells me that ISPCon speakers also stood up their audiences. At least at Seybold, the speakers came and met expectations.If you’re counting, that’ s two out of four trade shows that made the grade. Fifty percent is an unbelievable average for hitting a baseball, but it’s lousy for everything else.

In general, trade shows are going downhill. Some of it is the sour economy. Almost no major vendors, for example, thought it worth their time and money to have a booth at Spring Comdex.At the end of it, I’m left asking: What the heck is the point of going to trade shows these days?

Networking and making deals have always been the best reason to go. If you want to do that, you’re better off going to small, specialized shows like DTR’s. You’ll have a much better chance of finding someone who actually wants to do business.

People also used to go to learn about technology and business. But now it looks like you’vve only got about a two-in-four chance to see the “experts.” Frankly, I’m surprised conference attendees at Spring Comdex weren’t demanding their money back.

Another argument was that you went to trade shows to see the latest in technology. Forget that. If you want to see the best, you’re better off reading this magazine, eWeek or PC Magazine. That way, you get your information all nicely prepackaged, and you don’t have to pay $500 in airfare for the privilege.

First published in Ziff-Davis‘ Sm@rt Partner

February 27, 2001
by sjvn01
0 comments

The RIAA’s billion dollar blunder

As everyone who gives darn about music and technology knows, Napster offered the music industry a billion bucks over the next five years if they would, pretty please with sugar on top, not sue them out of existence. The music business, represented by the Recording Industry Association of America (RIAA) replied, “Hell no!”

Well, actually they used other words, but it doesn’t take an English professor to figure out their real message. There were catcalls saying that Napster’s announcement was only a public relations move, without any business plan merit. But what the recording companies really wanted was nothing less than to put Napster out of business, while whacking turncoat recording publisher and Napster-supporter Bertelsmann along the way.

Of course, the recording industry is right about problems with Napster plans. What they don’t get though, is that it’s also the last, best deal, they’re ever going to see.

The RIAA is acting like Napster, the company, is the enemy. It’s not. It’s also not Napster’s peer-to-peer technology, although that plays a much larger role. The recording industry’s real enemy is its own stuck-in-the-80s attitude. They can’t see that their old business model is dying and that to thrive in the 2000s, they must embrace digital music transfer technology. Peer-to-peer plus the Internet means a musical distribution revolution as big as the shift from LPs to CDs. Heck, what the music biz big shots should be doing is buying into Napster and co-opting it as fast as they can.

Why? It’s simple. Let’s say Napster, the company, goes down for good sometime soon. Will that stop people from trading music on the Net? No, it won’t even stop Napster, the program.

The genie is out of the bottle. Napster-style programs that use non-Napster company registered servers such as Napigator and Opennap Network are already running. Turn off the Napster servers, and the technically adept won’t even notice the switchover. It will take the rest of the Napster user community about two weeks. Similar peer-to-peer file sharing programs like The Freenet Project and Gnutella and its descendents BearShare, Gnotella and LimeWire will also continue to let people trade MP3 files with complete anonymity.

And, listen up RIAA; none of these new programs rely on centralised servers, management or even a company. When no one’s in charge, there’s no one to serve cease and desist orders to, no network of servers to be shut down. Here’s the important part: with no Napster, there’s no one left to make a deal with. Instead, there will be a dozen or more programs thinking around undercutting your old business plan even more effectively than Napster did.

In theory the RIAA companies are trying to create their own tamper-proof music exchange system with the Secure Digital Music Initiative (SDMI). But there’s a bunch of problems with this. First, the RIAA doesn’t really know how to go about implementing such a system. If they did, Napster would have never stood a chance in the first place. Secondly, the secure system will be quickly broken and the songs will soon show up on Opennap and Gnutella. And finally, the companies show few signs of having a business plan to put up their own music network in the first place.

No, all in all, the RIAA would like to pretend that the only way music can be distributed is on tape and CDs. It’s too bad for them those days are as over as the days of 8-track tapes.

What they should be doing is making the new digital music transfer age work for them. And to do that, they need Napster to be healthy, strong and part of their industry. With Napster providing the brand recognition and easily the world’s most popular song trading software, they could make the Napster billions and more. Bertelsmann understands that the music industry is profoundly changing, while the rest of the recording industry big wheels haven’t and that, within less than five years, will spell their doom.

A version of this story was first published in ZDNet.

February 3, 2001
by sjvn01
0 comments

Linux 2.4 is here, but some goodies are missing

Although the new bells and whistles are great, technology alone won’t guarantee success.

OK, Linux 2.4 can do serious enterprise computing lifting. With it, you can run up to 32 Pentium-class processors using symmetric multiprocessing (SMP). Come the day Itanium is really released, Linux will be the first operating system up and running on it, and youâ??ll be able to scale Itanium SMPs up to 64 processors.

The networking layer also is much more scalable, thanks to “wake-one” scheduling. That ensures that only one processor, instead of a gaggle of them, is listening for network traffic on one socket. Linux 2.4 has much better memory support, as well. Even 32-bit Intel processors running Linux 2.4 can address up to 64GB of memory. Thatâ??s not too shabby for all but the most-demanding database applications.

Some technical goodies are still absent. A journaling file system (JFS) is a necessity for an enterprise OS, and 2.4 doesnâ??t have one. But thatâ??s a minor detail. Come 2.4.1, which probably will be out in February, youâ??ll be able to easily deploy the ReisterFS JFS.

So whatâ??s Linux 2.4â??s problem? While the Linux network operating system is second only to the Windows 2000/NT family on Intel processors, according to IDC, to make the next step forward, Linux needs nontechnology goodies to convince CIOs to take the Linux plunge.

Hereâ??s what I think Linux needs to enable you to make a real run for contracts currently going to W2K/NT integrators. First, you need to educate decision makers about Linux and open source. Too many senior IT folks still think of open source as this scary thing that may come back and bite them with licensing problems.

Next, Linux value-added distributors and ISVs need to stick to open source and the Linux Standard Base. If Linux starts forking, itâ??s back to the Unix wars of the last two decades, and the only winner there was Microsoft.

Lately Iâ??ve begun to see signs that Linux vendors are getting more interested in proprietary Linux product add-ons. If that continues, we could see attempts to create proprietary Linux kernel programs and then Linux will be in a world of hurt.

Linux desperately needs serious management tools. Here, progress is being made. Caldera Systemsâ?? Volution looks pretty darn good for managing multiple servers, in my tests. Linuxcareâ??s Linux Managed Services also looks like an excellent service play. Finally, there are too many Linux certifications.

Red Hat has one, the Linux Professional Institute has one, and several training companies are pushing out other certifications. Linux needs one solid certification path that will be as well-known as an MCSE or CCIE.

If it gets all of those pieces into place, Linux will step beyond being a big business to become a big business.

A version of this story was first published in Ziff-Davis’ Sm@rt Partner.

January 8, 2001
by sjvn01
0 comments

Linux 2.4: It’s Here!

Latest and long awaited kernel is finally here and it will please the enthusiasts, but don’t rush to install on to your business servers.

It’s here! It’s here! Linux 2.4 has arrived and Linux fans are beside themselves with joy. If you have friends who love Linux you can forget about seeing them for the next week. They’ll be busy overloading broadband Internet connections downloading the new 2.4.0 kernel and compiling it.

Yes, it’s a great day for Linux enthusiasts. For users with personal systems, Linux 2.4 brings more device and Universal Serial Bus support. For administrators, Linux 2.4’s big news is that it makes Linux much more scalable. This going-on-a-year-late operating system is the one that should fulfill every Linux enthusiast’s–not to mention IBM’s–dream of Linux being a full-fledged enterprise operating system.

It just isn’t going to happen today — or this month — or this quarter.

Don’t get me wrong. I’m as pleased as punch that Linux 2.4 is here. I know where I’m going to be spending my time over the next few days. I’m going to be bringing my test bed Caldera 2.4 beta machine and two other test machines to 2.4 status.

What I’m not doing? And neither should businesses is taking my production Linux systems to 2.4. My VA Linux workstation running Red Hat 6.2 and my Corel PC are sticking with 2.2.x kernels. And on my production server I’m sticking with Caldera’s OpenLinux eServer 2.3 for now.

Why? Because, even after thousands, no, tens of thousands of programmers and beta testers have worked on 2.4, I know there will be bugs. Heck, I can read the Linux developers lists as well as anyone, I know there are bugs.

And, frankly, I don’t upgrade to any .0 release for exactly that reason.

For now, Linux 2.4 is for the adventuresome, the Linux lover — not someone whose bottom line depends on Linux. Besides, even after Linux 2.4 ages well, most users won’t get that much more from it anyway. Yes, Linux 2.4’s new device support is a win. But if you don’t have any of these devices on your PC, you’re not going to see much of an improvement.

So what’s the big deal then? The big deal is that Linux’s vastly improved support for clustering, multiple processors and memory are great news — for enterprise level servers. If you’re dealing with a basic Intel box as either workstation or server, Linux 2.4 isn’t that big a deal.

And if you do have top servers, you need more than just a high end kernel. You need programs that support the kernel’s new functionality, and those aren’t here yet.

Besides, if you’re serious about business Linux, you need the support of serious Linux vendors. Except for SuSE, they — Caldera Systems, Red Hat and TurboLinux–aren’t going to be releasing 2.4 kernels for their commercial products for a while yet.

To put it in terms of Microsoft operating systems, for most PC Linux users, the jump from a Linux 2.2.x to a 2.4 kernel will be similar to that from Windows 98 to Windows 98 Second Edition. Good, but no reason to spend the next 48 hours installing it unless you love the stuff with the passion. For server use, the Linux jump is much bigger; it’s like the leap from NT to W2K Advanced Server. But like Advanced Server, to get the most from it, you need the applications and support in place before making the jump.

The final answer is that while it’s exciting news for technologists, the Linux excitement for business deployments is still months away.

A version of this story was first published in Ziff Davis’ Sm@rt Partner.