Stoic Joker

Right, I was looking at duplicating the path on the (server) target machine also...which I'd assumed would also be part of their objective. Winbolic Link is another (junction point) 3rd party utility that it seems they don't want to use... *Shrug* ...This sounds like a job for DFS, but I'm not really sure what the "Sticking Point" in the client side D: -vs server side D: folder structure is. We may all be over thinking this.

He already has an existing D: drive so subst would cause a drive letter conflict resulting in one of them to disappearing. If it must be something native (barring the (free) junction point utilities), there is always there is always the mount a partition to a folder option. Create a folder called data on the existing D: drive, mount a partition to that folder, and put the data there. Just a Thought Stoic Joker

Um... Honestly, this sounds like an exercise in the the right way to do something wrong. Using the Server as a client endpoint of a VPN? Get a 2nd VPN box and setup a (proper) Gateway-to-Gateway VPN and 99% of the problems/headachs/etc. you're about to create will simply vanish.

Is there a screensaver trying to run in the remote session? How long does the session run properly (if at all) before it starts to lag? When the session is lagging, are the boxes other services also slowed? More detail = better answers.

Well... That tends to force them into an "All-or-None" situation, requires quite a bit of (Guaranteed) bandwidth, and tons of server side RAM. Are all of the users going to be on the road, and need full access to everything? Outlook can gain access to the Exchange server using RPC over HTTP to handle their Email/Calendar/etc. needs, and works quite well with (cellular broadband) AirCard connections. The finance application (What ever it may be...) may require TS to allow its on the road use ... But I'd try to keep some options open in regard to the other stuff to avoid completely locking people out of the office during an outage (...IT happens/Single point of failure stuff). I'm just thinking out loud here Stoic Joker

Boot to the recovery console of Win2k/XP CD and run fixmbr Win9x ver (run from startup disk) fdisk /mbr

Please tell me you're not planning on trying to run PhotoShop via Terminal Services ... There's not a right to do that. What is it you feel you need a terminal server for?

Okay... There's a couple of things troubling me at this point. First of which is what "Login Screen" is Atari37 getting when attempting to acess the FTP? A standard Windows login dialog? or The FTP based "Would you like to login anonymously" dialog. Reguardless of whether the FTP is running on a DC or a member server the authentication should be handled by AD, I've never seen a LM account work inside of a Domain environment and I administer several IIS FTPs as part of my daily activities. If adding the domain (Which should never be needed) information does allow access, is Atari37 running the FTP server in a DMZ and exposing the local file shares?!? IE's FTP access can be handy in a pinch... but for proper testing, one should really be using a real FTP client application to be sure they're connecting to port 21 (and only port 21). One thing to check would be what is atari37 typing into the browser, if it's only the IP address (or host/DNS name) of the server then they'll get a HTTP connection (and standard Windows login dialog) by default which will require the domain name to be added for authentication. In order to connect to the FTP they would need to specify it in the link used in the address bar e.g. FTP://servername ...That (and only that) would force the browser to connect to port 21 and access the FTP server...which I suspect at this point to be the real issue.

That depends on how he answered the Isolate Users option during the setup of the FTP site as there are 3 options: 1. Don't isolate users. 2. Isolate users to their home directories (I'll assume this is the local accounts only part you spoke of) 3. Isolate users to their home directory using Active Directory.

C/C++ is my primary choice also.

Really? I found several, But I really liked this one: I haven't experienced the problem personally, but hopefully this will help.

If you Like DOS: COPY /B file.PRN LPT1 If you like GUI's http://www.magma.ca/~russrite/PrnPrint/index.html

Visual Basic?!? (Icky) I'd rather be horse whipped ... If he can already do basic C then go for (Pure Win32 API) C++, then you end up with tight fast code that you can drop anywhere and know it'll run. C# is nice if you need to do some thing large, complex, and in a hurry. But you're saddled with the (ever version specific) .NET runtime requirements (and its accompanying 50MB download) just to get the thing to run. You'll only try to do something quickly by tossing it on a network share, and running it on all workstations once before you realize what a pain C# can be. A. The .NET 3.0 (or 2.0) run-times are not a critical update ... so nobody will have them installed. B. The program won't run because it doesn't have permission to do so (e.g. code level security is built into the .NET run-times). C. The exact syntax to tell the %$^%^&% think it's OK to execute isn't real easy to remember Mainly because you're right in the middle of trying to get something else done ... That you were trying to do quickly for a reason... I used C# once to get something developed in a hurry, and it worked beautifully... Deployment however (see above...) went to hell in a handcart. I will never code in anything else but (completely runtime-less) Pure Win32 API C++ code ever again!

If MAC filtering isn't good enough, you're looking at 802.1x and a RADIUS server if you want additional security (802.1x will basically only allow EAPOL packets from a machine before it auths against the RADIUS server). I could be way off base here...but desn't the 802.1x/RADIUS combo assume wireless access control? If sombody wandering about the building could toss a patch cable (we all carry them) into an open hardwired port and get an IP ...No? While not as fancy, why not try using DHCP User or Vendor Class I'm just thinking out loud here Stoic Joker

LOL Well...I'm in the same (MCSA 2k) boat, so good luck to both of us! Can you give some details on how this DHCP lockdown requirement came about? How often is it causing how much of a problem etc.? Static addressing for 1000+ machines has got to be a nightmare.

http://www.codeproject.com/ http://www.planetsourcecode.com/ http://www.programmersheaven.com/

What type/level of access to the machines in VA do you need? Shares? RDC? The VPN should handle the routing for you e.g. any packet for target network X will be stuffed into the tunnel and pop out "there" (Tis their nature). So if the 5 machines in VA are statically addressed you could just use their IPs. Or setup a WINS server in FL and have them register with it.

All machines in my lab are on 24/7. Primary WS and servers are on battery backups, and I have blown the dust out of a machine while it was running more than once.

Well it's kinda clumsy, but assuming your MAC reservations are sequential you could do an address exclusion for the unreserved block of addresses remaining in the scope. btw I noticed your member title says MCSA W2k ... Aren't they (MS) retiring the MCSA 2k3 upgrade exam (70-292) the end of this year?

On the RDC configuration dialog, go to the Local Resources tab and at the bottom under Local Devices select Printers. Then when you connect to the office the local printer will show up as an option on the remote desktop. How fast it will print depends on available bandwidth and how neatly the printers driver handles a print job. A business class laser printer with a driver that just prints will be (flawlessly) quite fast. On the other hand if you have a (typical) Ink-jet that likes to send volumes of bidirectional data back and forth about every move the printer is/will or might make ... then chances are the print job will fail (assuming the print driver can actually manage to shut-up long enough to get wedged into the remote session. Many of the better printer manufacturers will have two versions of their printer drivers available: The "Home User" Natters like a Jewish Mother garbage version. and the Corporate stick to bloody business and just print version. I strongly recommend picking the 2nd (corporate) version just because it far less likely to crash your computer with some idiotic spool hanging eye-candy widget that some clod in marketing thought might entertain the user.

Doctors Office = HIPAA Laws = You need a VPN between the 2 offices. If the practice software is going to be run via Terminal Services, then the client machines don't actually have to be part of the domain. Policy management is simpler if they're joined to the domain ... but TS doesn't require it. The TS is installed via ad remove windows components (or use the server wizard) and will need to be activated by MS. Also I hope you budgeted for TS licenses because they're (not cheap) extra. P.S. Don't double post ... It tends to tick people off

The flip side of that being that half of the stuff being built today is designed to be disposable, there's no point in fixing something that you're supposed to toss in the garbage (Ah the Joy of Planed Obsolescence). However as a Microsoft Certified Partner, I will agree, their Tech Support is Excellent. One guy (from Canada) was walking me through an Exchange recovery (power outage/corrupted mail store) while walking his dog ... and he never missed a beat. We had the box back online in less than an hour. So not all out sourcing is bad.

The "Original" (LM only) version that was on page one of the thread that day (on XP). I'm running Vista x64 Business Edition here in my home office, and am getting the following error when trying to run Key Decoder: Component 'comdlg32.ocx' or one of its dependencies not correctly registered: a file is missing or invalid. Note: All my dialogs Common and otherwise are running fine ... so I'm guessing it's the proggie. (But Seriously...) I don't recall if it was the header file (.h), the library file (.lib) or both that have/require updated versions to resolve the error (in C++), but I've run into it before.

Well ... with your title of "The Hardware Guy" ... It makes one hesitant to answer. But if nothing has changes, and it ran before, it kinda sounds like a hardware issue. Especially with it not having any other events & it not being able to BSOD. I'm thinking either Memory, PSU, or Dust (Heat/Static "short"). Have you checked the (Captain) Obvious stuff? Clean & re-seat memory, blow the dust-bunnys out of the case, back-off any overclocking, etc.? Don't Shoot Me ... I'm just thinking out loud here. Stoic Joker

LOL I was about to post the same link but you beat me to it. The site has been there since some time in the late 90s (that I know of...) But it is an excellent tutorial on Subnetting the I recommend frequently.