RogueSpear

If pinging an IP address isn't working then there's serious issues (unless a software based firewall is blocking ICMP traffic). Make sure that both computers are on the same subnet and that both have the same default gateway. Once pinging works for IP addresses, then ping using host name to see if DNS is functioning properly.

This may be overkill to some people, but whenever I install a piece of software for the first time I always install it on VMware first. And usually when I do that, I'm also performing a two step snaphot using AdminStudio so that I know every single file and every single registry entry made. Call me paranoid, but when I'm preparing to install an application to over 100 workstations, I want to know what's going on. And I've been burned pretty bad in the past too. I don't want to repeat that experience.

I'm on Cingular with an unlimited data plan. Using a Sony S710a (Class 10 EDGE) via Bluetooth. In my observations I have a satisfactory web browsing experience, but it comes with some caveats. The connection seems very sensitive to multiple connections. What I mean by this is if you go to a web page that has to download tons of images, it will slow down significantly. If the page is simpler, even if it has higher resolution images, so long as there aren't too many of them, it "feels" fast. This seems to jive with what chilifrei64 was talking about with "burstable" connections. It seems analogous to doing a file copy. If you copy a subdirectory with 5,000 files totaling 1GB, it will take a lot longer than copying one single 1GB file. I had some major issues at one point with my cable modem (and the cable company) where I was without my broadband connection for close to two weeks. During that time my cell phone was my link to the internet. I did everything on it too including VPNing in to work daily. I have to say that it wasn't bad at all. In fact I'm still considering dropping my $50/month cable modem completely and just relying on the $24/month I pay already to Cingular.

What's rather odd is that it's very possible to install Adobe After Effects 7.0 and Adobe Encore 2.0 without all that extra stuff. I can't wait to see what they do with all the Macromedia assets they recently aquired (said with sarcasm).

Jeez, it seems so obvious now it's almost embarassing EDIT: OMG, I still can't find any documentation regarding this over at MSDN. Either I'm getting brain freeze or it's really hidden.

I always try to pay close attention to the contents of .inf files whenever I come across them, be it part of a driver package or part of a software install, and I've never seen an example where the .inf file itself initiated the execution. The closest that I have ever seen is setting the registry entries for RunOnceEx or something similar. I also did a lot of reading over at MSDN regarding .inf file programming, but was unable to put anything together. Of course part of that may be because reading MSDN brings about narcolepsy

Every 3rd party firewall I have tried sucks in it's own way. It's really pathetic the garbage out there. Believe it or not the only firewall that seems to work problem free for me it XP's built-in firewall. And yes, I'm aware that it's unidirectional. Every other firewall out there needs to be disabled way to often for one reason or another.

Got DP Modem and TV up. Just have DP Monitor left.

I couldn't agree more

We'll probably find out in another thread sometime in the future

To start, with both sides of the network sharing the same subet, all the prior settings are pretty much invalid, with the exception of possibly your DNS server settings on the server (how's that for a bad sentence). Here's what I would do, I'm sure others would disagree. Turn off DHCP serving on your SMC router. SBS should provide that. It's not really a good practice to have a server obtain it's network config via DHCP. Also, SBS will see that there is another DHCP server and flag it as a rogue DHCP server. Then there's the issue of your clients sending out a DHCP request and getting multiple responses. They'll take the config of whoever answers first. Could be SMC, could be SBS. Bond the two interfaces on the SBS server into a 200Mbs virtual NIC. Take note that when you do this, the Intel drivers may create a MAC address for the NIC that is entirely different from either of it's "real" MAC addresses. Or it may let you choose which real MAC of the two you want to use. I've done this on Compaq Proliant servers and on Dells that uses Broadcom NICs, but not on an Intel card. In the NIC's properties, set a static IP and the default gateway as the SMC. Enable SBS to serve both DHCP and DNS. Make your forward lookup zone an Active Directory integrated zone. For the DHCP scope, configure it so that your clients point to SBS for DNS and to the SMC for the default gateway. There are a few settings both in DHCP and DNS that are optional, there are some that are good for security, etc. Once you get things up and running, you can read up on what some of them do and set appropriately.

Perhaps KtD based on interface (USB/Firewire/PCMCIA) ? This is something I've been toying with now for a little while, but I'm hesitant to devote any real time to it because I don't know if you plan on implementing this or not.

Without being able to look at everything, I can only guess. Perhaps the non-admin user doesn't even have the permissions to read the Default User profile. Did you accidentally create the new user as a Guest instead of a User?

Why don't you just import it with the rest of your reg tweaks. regedit /s tweaks.reg

I notated all the updates in the post. The only point of confusion perhaps is that I essentially eliminated DP Notebook. The alps touchpad is in DP HID and the battery is in DP Misc. I did recompress everything using max compression.

Well some of them come with no MAC address apparantly and I don't think I've seen a single success report of someone being able to perform a RIS boot with one. Perhaps this is why they include a NIC from another manufacturer.

[Offtopic] A few months ago I ditched Symantec Client Security, after having been a customer of 125+ seats for over three years because it just plain old didn't work as advertised. The final straw was it's incompatibility with Gmail. Whenever I wanted to access Gmail through a web interface, which is pretty much whenever I'm on a computer, I always had to disable the firewall. Real secure wouldn't you say? Well I've been testing out all of the other major players out there and I have to say I'm not impressed in the least. They all interfere with something or don't really work as you'd expect them to. Guess what I've been using problem free for the last two weeks? Windows XP firewall. I know it doesn't block anything going out, but I figure it's better than no firewall at all. And for me it's way more secure than any other firewall out there because I can actually leave the dang thing enabled all the time! Seems like the more time goes on the worse things get when it comes to trying to protect yourself. [/Offtopic] Anyway, I can only imagine the difficulties that a firewall from a company known for it's great NICs and firewall expertise like nVidia would give you.

Well I'll try without getting into a full blown lesson on TCP/IP. The simple explanation is that a subnet is the local network that traffic does not need a router to access. Everything within a subnet will receive broadcast traffic from a node on the same subnet. I'm not the best at outlining these things, so maybe using your network as an example will help. What it sounds as though you want to do is have all of your workstations on one subnet (192.168.2.0 as an example) and have your router on a seperate subnet (192.168.3.0 to continue the example). Your server is acting as a router, seperating the traffic between the two. At least this is how it appeared to me from your diagram a few posts up. The benefit to this arrangement, and this is a matter of debate, is that should someone compromise your router, they should only be able to directly attack your server and not the workstations. This is because the server sits in front of the workstations. Now the server is arguably the most important node on the network, so having someone attack it instead of a workstation is where the debate comes in. There are other factors. If you're not hosting a web site accessible to the public or some other service where someone from the outside is supposed to gain some kind of access to the inside, then a properly configured router will sufficiently keep intruders out (not counting downloaded trojans, etc). I'm going to go on the assumption that your SMC router is a consumer grade router with limited configurability and interface options, as opposed to something like a Cisco PIX 515. So in all actuality, I think you'd be well servered to just stick with one subnet and bond your dual interface NIC into a virtual NIC. This will cut down on the complexity and eliminate some possible performance bottlenecks. Also depending on how things are configured, it will eliminate a potential point of failure. If you can manage to setup and configure a VPN setup using Windows native VPN services, then this should be a walk in the park for you.

Boardbabe outlined it perfect. The only thing I would add is that your directory "Apps" will likely be C:\Windows\Apps (assuming that you're using drive C and that your using the default systemroot name of Windows). Anyway, I just wanted to point out that that's where Apps will wind up in case you didn't mean for it to be there. It's fine if it is there, just not where I would have thought of putting it is all.

The default gateway in the scope should be 192.168.2.1 (the router), not 192.168.2.108 (the server). The other suggestion I would make here is to disable the router from offering DHCP leases since your SBS server is acting as a DHCP server on the same subnet. For starters the SMC router will show up as a rogue DHCP server on the SBS server, secondly who know which DHCP server your clients are getting leases from. Considering everything here is on the 192.168.2.0 network, you would probably be better served to just bond both NIC interfaces on the server into a single virtual 200Mbs port. EDIT: Ok, I just noticed in a post subsequent to your first post, that the server sits between the router and your clients. If you plan to keep that arrangement, you should use a different subnet for each side of your network.

At the top of post #6 in this thread. Tested the link and it's working fine.

http://www.msfn.org/board/index.php?s=&sho...ndpost&p=424509 I tested all the links and they seem to be working. Let me know what you guys think so far..

This method should work fine for most purposes. I never checked to see if it actually changed the reading in the dialog box in the screen shot above, but I have to do this whenever I install a proprietary application for one of my clients. The installer will not install on XP Gold or XP SP2. It wants XP SP1

Here is the line to compress all of the log files: 7za a %systemroot%\LogFiles.7z %systemroot%\*.log I like to just leave it there so I know where it is, but obviously you can plop it wherever you like.

Something to keep in mind is that if you place something in \$OEM$ it does get copied to the hard drive, but only temporarily. For instance, in \$OEM$ you might have your cmdlines.txt file and a file called tweaks.reg. In cmdlines.txt you call for tweaks.reg to get imported. Now you wouldn't expect to find cmdlines.txt and tweaks.reg in the root of your system drive after setup has completed. Well it's the same with anything else you put in \$OEM$. If you do in fact want something copied and to stay there after setup, then put it inside $OEM$\$1.