OSNews

It's up. Check here for all of the updates:
http://www.microsoft.com/technet/windowsserver/sp2.mspx

You could test it....

Testing is always a good idea. In an ideal universe, you have one server standing by, identically configured to one of your production servers, and you apply the service pack on that, then push out the updates to your other systems a couple days later, assuming it didn't bork anything.

If you're fortunate enough to have a VM cluster, this is fairly easy. If you're not, having the extra hardware lying around for the sole purpose of being a guinea pig is difficult to manage-- Note that the hardware ought to be as similar as possible to the production server, or you're going to find out which device drivers aren't working.

Short version: It's rarely an ideal universe.

I was under the strong impression that most Server Hardware uses very standard devices that would come with drivers on the Windows install media or Windows Update. I definitely wouldn't use non-WHQL drivers on a server.

It's not just about hardware, This SP adds/changes signifigant functionality, and I would want to make sure my apps don't break

Not to mention that you should have been testing the pre-release versions (betas and RC's), all of which were available to anyone willing to spend the bandwidth downloading them.

From our own internal testing, we're seeing SQL Server 2005 performance increases of almost 20% in some cases, which is the main selling point for development environments.

That would be great if it held up. Our system could really use a boost from SQL operating more efficiently.

What would you have them do? Would you prefer no patch at all? We'd then see a comment complaining that Microsoft haven't fixed problems.

Take a look at any operating system over the past ten years and patches and updates are present.

Get a grip.

If you are running such important servers, I would thjink you would have a test environment. So test the SP, make sure it doesn't break anything, fix anything it does, and then deploy. Windows SP's are no different than any other OS update by any other vendor or distro

Its so much easier to come on OSNews and troll, complain and make up misinformation than it is to go out and test the patch. I am actually testing it myself and so far no problems.

I haven't had any problems either, two days, no issues, might be able to deploy next week, hopefully

R2 is a seperate release that includes additional features like the Subsystem for UNIX-based Applications, Common Log File System, and AD Federation Services.
http://technet2.microsoft.com/WindowsServer/en/library/f9d70026-ae8...

SP2 does work on R2.

Well... it's for server 2003, not XP Pro.

It's for XP Pro **x64 Edition** too.

Why would a server service pack have the ability to install on a client OS? That's just weird. What can you possibly look to gain from that?

XP64 is based on win2k3 code, so it's more 2003 than XP

That still doesn't explain the desired end result.

Windows XP x64 is based almost entirely off of the Windows 2003 x64 code base. It's also clearly labeled on the MS downloads site that it's for Xp x64 as well.

corrupt profile possibly?

I'd create a new administrator profile and try installing the update again.

Common MS issue it seems, profile corruption that is.

i just clieked ok to the error, and the installation proceeded as normal, after the reboot it was SP2, but i did note the following:-

windows live messenger started an MSI as soon as i logged in, which stated that it was 'repairing itself', so I guess the two could have been related.

on another note, now RIS doesnt work at all, however I cannot identify yet if that is because I changed IP address of one of the network cards on the server (the internet facing one, not the RIS service nic) and there a re 1001 DNS errors as a result. I'll play around with the server tonight and see can I fix it.

cheers
anyweb

They did. It's called Windows XP Professional x64 Edition.

Microsoft always has a habit of taking a great operating system and royally rooting it to kingdom come.

Case in point, NT was a rock solid, be it slight slow due to hardware constraints, micro-kernel based operating system, very small, compact rock solid kernel with security built into it from the ground up.

Everything came unstuck when management wanted the graphics more snappy by pushing it into the kernel, wanted compatibility put ahead of implementing things properly.

Fastforward to today, take Windows 2003 SP1 and stuff it up by adding layers of compatibility and features; the new features are great, but quite frankly, the backwards compatibility should have been thrown out - yes, up take would be slow as companies would have to stop using removed API's like GDI/GDI+ for example, but at the same time, the need to have layers of abstraction and compatibility would reduce the bloatage and force good coding practices on third party software vendors.

Back on topic, it'll be interesting to see how Windows 2007 Server will be like compared to Server 2003; from what I understand, the next 9 months or so are being spent doing some major overhauls in Windows, but I am skeptical as to whether they'll put quality above meeting deadlines.

When you take into account a rejuvenated Sun Microsystems with development on Solaris x86 going gang busters, Linux vendors pushing out enterprise server/workstation upgrades or updates (Red Hat Enterprise 5.0 and SuSE Linux Enterprise Desktop SP1), Microsoft can't afford to be making the same sorts of mistakes they did in the past as they know that there are viable alternatives, and both regulatory bodies are keeping a close eye of Microsoft.