DistributionWatch: SCO Linux 4--Ready for the Big Time - page 3
Down to Basics
To make SCO Linux 4 do its stuff, you'll need at least an Intel 486, with 64MBs of RAM and 500 MBs of disk space to give it a try. But, that's pointless. To do the jobs SCO Linux 4 is meant to do you'd need a minimum of a high-speed AMD Athlon or Intel Xeon with 512MBs of RAM and 40GBs of hard disk and up.
Since I don't have one of those in the office (darn it!), I tested SCO Linux 4 on a HP Pavilion 512N with a 1.4Ghz AMD Athlon XP processor with 512MBs of RAM and an 80GB hard drive. By UnitedLinux standards, that's barely getting into second gear.
Even so, some things quickly became apparent. One is that SCO Linux 4 is easy-I mean fall off a log easy-to set up. With more two decades of setting up server operating systems under my belt, I've never seen one this easy to set up before. In fact, I've found most desktop systems to be more difficult to install.
In large part that was because SCO's Webmin and Usermin, Web-based administration programs are very easy to use. We also found, though, that YaST, the UnitedLinux default administration suite, also worked well.
For fine-tuning, though, they weren't perfect. Both use KDE 3's built-in Web browser, Konqueror 3.03, for their interface. And, I found that Konqueror consistently broke during some setup installations. For example, it always broke during some stages of setting up Samba, the Windows NT compatible file server. I was able to get around this by using Samba's own SWAT administration tool. From some early experiments with Mozilla 1.01, the other supplied Web browser, it would appear that it works more reliable with the Webmin and Usermin administration tools.
While I didn't test performance as such, I did run some informal tests of how fast it ran compared to Caldera's pre-UnitedLinux Linux, OpenLinux 2.4. I found that on the exact same machine, SCO Linux 4 and its applications ran faster.
And, for lack of a better term, it ran smoother than its predecessor and other Linux distributions. There were fewer glitches. Yes, Linux is more stable than its competitors, but we all know programs that need fine-tuning before they work well enough for business. Well, on SCO Linux 4, there were simply fewer fit and polish problems.
That's pretty amazing for a 1.0 release. Of course, if you look really closely, you can see a lot of bits and pieces still labeled SuSE rather than SCO or UnitedLinux, but for practical purposes of getting the job done, it runs remarkablly well for a 1.0 release.
Still, nothing is perfect and neither is SCO Linux 4. The biggest problem I found was that there is no graceful way to upgrade from OpenLinux 2.3. In talking with SCO, I discovered that it wasn't just my own klutziness getting in the way. The only way to 'upgrade' OpenLinux, or any other Linux for that matter, is to back up your data and configuration files and restored them after letting SCO Linux 4 blow away the existing Linux system. If you've invested a lot of time in getting your Linux setup just so, be ready to re-do a lot of it.
So, in short, if you're upgrading an existing business installation, make sure you have up to the second backups and allow for lots of time for bring the system back up to production level. Otherwise, you're going to have one really ticked up client on your hands.
UnitedLinux representatives tell me that upgrade paths from older Linuxes will be made cleaner. But, with the possible exception of SLES users, I doubt that will happen. By its very nature UnitedLinux resets all those little, but vital, Linux file placements and settings to one standard way. And, that way, again with the exception of some of SuSE's Linuxes, isn't the UnitedLinux way. In the future, however, one SCO Linux 4 is in place upgrading SCO Linux Upgrade service, will go much easier.
Solid state disks (SSDs) made a splash in consumer technology, and now the technology has its eyes on the enterprise storage market. Download this eBook to see what SSDs can do for your infrastructure and review the pros and cons of this potentially game-changing storage technology.