Tech Support Section Home Mekong Network Home
About This Site...
Mekong.Net Sites...
Recently Updated...
Recommended Links...
Contact Us...
Questions? Comments? Requests? Click here to contact us.

Infrequently Asked Questions

I've labelled this page "Frequently Asked Questions" on the links, but the title above is probably more accurate. Who the hell asks questions like these, really?? Has anybody other than me ever asked? I dunno. Still, just for the hell of it, here are a few poorly organized notes on this that have baffled me over the years.



Server 2008

Where do I set the domain name suffix on a stand-alone 2008 server?
Some server features require the use of a fully-qualified DNS name. But what if your server isn't a member of an Active Directory domain? To set the DNS suffix, go to the System Properties, click the button next to the line that says "To rename this computer or change its domain or workgroup, click Change." Then, on the screen for "Computer Name/Domain Changes," click the "More" button, and enter the desired domain in the field for "Primary DNS suffix of this computer."

How can I make a program start automatically for all users?
On Server 2008 (and Vista, and Windows 7) the location of the "All Users" startup folder has changed. In the past, it was typically at "C:\Documents and Settings\All Users\Start Menu\Programs\Startup". Now, it's at "C:\ProgramData\Microsoft\Windows\Start Menu\Programs\Startup". (You'll need to have Explorer set to show hidden files in order to see this directory.) By putting the program (or a shortcut to the program) in this directory, it will automatically start when a user logs in.


Windows 7

I'm trying to install Windows 7, and it's very, very slow. REALLY slow. What's wrong?
The Windows 7 setup program can be confused by BIOS settings relating to floppy drives. In particular, if the BIOS is configured to define a particular type of floppy drive, and no drive is present, Setup will be horrifyingly slow. There's a discussion of this problem in a thread at In my case, the system was an Asus M3A, with an Athlon X2 dual-core processor. Setup was so slow, that I actually walked away from the machine thinking it had hung; but when I came back five minutes later the next screen had appeared. Tasks that should have taken seconds, however, took minutes. However, after rebooting and changing the BIOS settings to disable the floppy drive, setup ran fine. The thread referenced above, incidentally, indicates that this is also a problem on Vista.

How can I shut down Windows 7 without installing updates?
Pressing Alt + F4 will bring up the old-fashioned "What do you want the computer to do?" dialog, with a full array of options: such as Switch user, Log off, Restart, Sleep, Hibernate... or Shut down without installing updates. Another option is to use the command line: shutdown /s will initiate a shutdown, and it will ignore the updates.

Why can't I telnet on Windows 7, or on Windows Server 2008?
By default, a telnet client is not installed on Windows 7, Server 2008, or Server 2008 R2. However, you can easily enable it from the command line. Open a command prompt (with Administrator privileges) and type this command:

pkgmgr /iu:"TelnetClient"

You won't see any confirmation in the GUI, but once the command has completed, you'll be able to telnet from the command prompt.

How can I remove Service Pack uninstall files on Windows 7 or Server 2008 R2?
In earlier versions of Windows, you were given a choice to choose the location for storing the uninstall files whenever you applied an operating system service pack. That made it easy to free up disk space later on: you could just delete the entire uninstall directory. Now, however, Windows stores the backed-up files automatically, and it's not obvious how you can remove them. The files can, however, be removed using the Deployment Image Servicing and Management Tool. The command is:

dism /online /cleanup-image /spsuperseded

Incidentally, you have to specify: "spsuperseded," and not "spsuperceded." If you're like me, and prefer "superceded" to "superseded," well, it turns out that you're just plain wrong. According to Merriam-Webster, even though "superceded" has been in common usage since the 17th century, it's still regarded as a misspelling.

Where the hell is the "Send To" folder in Windows 7?
There were many interface changes is Windows 7 (and Vista) which seem to be designed to make it less convenient to use your computer. One such change was moving the location of the "Send To" folder, thereby making it harder to add time-saving shortcuts. If you're looking for it, try this location:




Why can't I delete or relabel a partition in DOS?
Suppose you're trying to relabel a partition in DOS. But you can't: you keep getting an error message saying "Cannot make directory entry." Or maybe you're trying to delete a partition using DOS FDISK, and you keep getting an error saying that the volume label does not match. What's wrong? Here's a guess: the disk was previously labelled using Windows 95 or later. You can't relabel a disk, either hard or floppy, with DOS 6.22 or earlier once it's been labelled with a later OS. If you don't need the data on the disk, you can get around this by using the FORMAT command with the /u option, such as:

format c: /u

If you DO need the data on the disk, I think you'll either need to relabel the disk in Win9x, or use a utility such as Norton Disk Editor.

Why can I sometimes not copy all the files in a directory from the command line in DOS or Windows?
The copy command in DOS and Windows 9x will not copy zero-length (empty) files. The copy command in NT and Windows 2000 will. I'm talking about FILES here, not folders. Why would you have an empty file? You might have a program or batch file that creates a file to log output, but then no output is actually generated. Presto... empty file.


Windows 95 and Windows 98

Where can I view a list of the hardware installed on my PC in Windows 98?
Windows 98 includes a hidden, unsupported tool for viewing hardware information. To launch it, enter this command from the "Run" dialog in the Start Menu:

hwinfo /ui

You'll see a list of all hardware devices in the PC, including driver files and registry entries related to each device. However, this officially "unsupported" tool isn't perfect: it will sometimes list errors for devices that are working just fine. Still, it's a good way to get a quick snapshot of the system's components.

Where can I see software version and system information in Windows 98? And how can I decide what does or does not load when my computer boots up?
In addition to the Hardware Info tool described above, there are two other tools in Windows 98 which can be used to see system information: the DirectX Diagnostic Tool (dxdiag.exe), and the Microsoft System Information tool (msinfo32.exe).You can launch either of these tools by typing them on the "Run" command line:


You can also launch the Microsoft System Information Tool from the Start Menu. It's usually at Start | Programs | Accessories | System Tools | System Information. You can open the DirectX tool from the tools menu of the MSI tool, also. But the best part of the MSI tool is the System Configuration Utility, also found under the tools menu. That utility allows you to set general configuration options, including what programs load automatically at startup. I deeply believe that the fewer things loading at startup, the more stable your system will be. Click the "Startup" tab within the SCU, and prune out all non-essentials. Sometimes it's not easy to tell what is necessary and what isn't, but with a little research and a little practice, you can learn.

How can I use system variables in Windows 95 and 98?
To see the list of current variables, type "set" at a command prompt, with no parameters. To set a variable, use the command:

set my_file_dir=c:\transfer

To use a given variable, precede and follow the variable name with the percent sign. For example, if you want to see the files in the directory you had defined as "my_file_dir" you could type this command:

dir %my_file_dir%

And if you just wanted to know what one specific variable was set to, you could use "echo", like this:

echo %my_file_dir%

Netscape just lost a bunch of my mail! What happened, and how do I fix it?
Netscape stores email in a file called "inbox." There is also a file called "inbox.snm." The second file is an index listing the contents of the actual "inbox" file. Sometimes this index becomes corrupted, and it will give an inaccurate listing of what is in the mailbox. Deleting this inbox.snm file will force Netscape to create a new index. These files are usually in "C:\Program Files\Netscape\users\(username)\Mail", or something similar. Make sure you close your mailbox before you try to do this. And, if you are the cautious sort (like me), it's better to rename the file, rather than just deleting it. That way, you can restore it if you find that you've just made things worse.

My modem makes a hideous screeching sound every time it connects. How can I get it to shut up?
Open the "Modem" control from the control panel. Highlight the modem in the list of installed modems and click the "Properties" button. Click the "Connectin" tab, and then click the "Advanced Properties" button. In the "Extra settings" field, type M0. (That's a zero, not a letter "o"!). This should set the modem speaker volume to zero.

I'm using Microsoft Word 97, and it's minimized on the task bar. But I can't restore it, or I can't see it when I maximize it. What's wrong?
It may be that the window is set to appear at coordinates that are offscreen. This can occur if you've reset the screen resolution. To fix it, you can delete the registry key that stores the coordinates. (Remember, editing the registry is risky, and you'll probably explode if you do something wrong. Do you really trust my advice? Check out the Microsoft KB article at Q264298) Delete the key:



Microsoft Outlook

I've got a new computer, and Outlook's autocomplete no longer knows the email addresses as I type them. Is there a way to retrieve the old autocomplete list?
If you still have the old computer, yes. The autocomplete name cache is kept in an ".NK2" file for the Outlook profile. Typically, if you have only one profile, the file will be named Outlook.NK2. Suppose your login name is TRoosevelt. The default location would be:

C:\Documents and Settings\TRoosevelt\Application Data\Microsoft\Outlook\Outlook.NK2

Make sure Outlook is closed, then copy this file and put it in the appropriate location on your new computer. When you open Outlook again, autocomplete should be working. A related note, however: this file wasn't designed to hold large numbers of addresses, and it will sometimes become corrupt. Supposedly, the maximum number of names it will hold is roughly 1,000. (See details at If it becomes corrupt, you might see all kinds of odd behavior, including inaccurate address completion. If you rename the file, Outlook will create a new, empty one the next time it starts.

I have a distribution list in Microsoft Outlook. In the past, if I changed one of the user's addresses in my address book, I could just click the "Update Now" button on the distribution list page, and it would resolve the new address. Now, however, it doesn't work. What's wrong?
This can happen if you've changed your configuration of Outlook. If, for example, you've upgraded from Outlook 98 to Outlook 2000, you might have moved the records from personal address book to a Contacts folder. Outlook will always look for distribution list updates in the location where the record originally resided. If you want to be able to update the records in a single location, you'll need to delete the old entries out of the list and re-import all of your address from your new Contacts folder.


Internet Explorer

All my toolbars just disappeared! What happened, and how do I fix it?
Later versions of Internet Explorer have a feature called "kiosk mode," which hides all the address bar and/or the toolbars. You can get in and out of kiosk mode by hitting the F11 key.

Every time I start the browser, it tries to go to, and fails with an error. How can I stop it from trying to launch runonce?
This is pretty annoying... even if you reset the home page, the browser still insists on trying to launch its idiotic "run once" routine, which (I think) is supposed to let you configure the browser settings. To make this go away, you need to have two registry keys correctly configured. They are both in HKEY_CURRENT_USER\Software\Microsoft\Internet Explorer\Main. The two keys are "RunOnceComplete" and "RunOnceHasShown". Both are DWORD values, and they should be set to hexadecimal value of 1. The computer where I saw this problem had the RunOnceHasShown key with the correct value of 00000001, but did not have the RunOnceComplete key.

I use Firefox as my default browser, but there are some sites I need to open in Internet Explorer. How can I make a shortcut to do that?
If you put a shortcut to a website on your desktop, and then open it, it will always open in the default browser. You could, of course, launch the other browser, and keep the site as a favorite (or even as your home page) but that's a pain in the butt. It would be easier to just double-click on the desktop shortcut.

To do this, you'll need to create a shortcut to your required browser, and then add the URL as a parameter to the shortcut target. Put a shortcut to the browser on the desktop, then right click on it and choose "Properties." Let's say the target of the shortcut is:

"C:\Program Files\Internet Explorer\iexplore.exe"

At the end of that target line, add a space and the URL. For example:


"C:\Program Files\Internet Explorer\iexplore.exe"

You might want to make one other little tweak. This method will give you the browser's icon for the shortcut. If you want to change it to the site's icon, you can usually download "favicon.ico" from the root of the website. Or you can convert an image from your own collection to a ".ico" file with some graphics programs, or with an online graphics converter like the Cool Utils Image Converter.

Windows Server 2003

I need to connect to a remote server via RDP, but it says that "Terminal Server has exceeded the maximum number of connections." What can I do?
This usually happens when users close RDP sessions without logging off. When Server 2003 is in Remote Administration mode, you're allowed one console session, and two RDP sessions. You've got a couple options. One way of getting around this is to connect to the console session. The console session gives you exactly what you would see if you were logged in directly on the machine. (It's also where popup errors will be displayed; if, for example, a service fails at startup, you won't see the error immediately on login with other Remote Desktop Protocol sessions. You'll only see it on the console.) One way to get the console session is to create a shortcut to RDP, and append the " /console" or "/admin" option to the shortcut's target. (If you are using Windows XP with Service Pack 3, it's the "/admin" option; for earlier versions, it's the "/console" option. A pox upon the jackass who decided that it was necessary to change that.) The behavior of the console session is different from other sessions in that if someone is already logged in on the console session, you'll still be able to connect, but will see that the computer is "locked." Assuming that you have admin rights on the machine, you'll be able to log them out, and log in under your own account.

Another option is to use the Terminal Services Management console. You can launch this by entering the command "tsadmin" on the Run line, or via a command prompt. Once the managment console opens, you can select "Connect to computer..." from the Action menu, and enter the name of the computer you need to access. Once you've connected, you'll see a list of all the RDP and console sessions, and you can disconnect users to free up a session.

How can I stop the computer from always asking me to supply a reason for shutting down?
By default, Server 2003 insists that you explain why you want to shut down or restart. That's annoying. To make it stop, you need to disable the Shutdown Event Tracker. That can be done by editing the appropriate group policy. If you only want to disable it on a single computer, you can launch the policy editor with the command:

Go to Computer Configuration, expand the Administrative Templates section, and then highlight "System." Double-click "Display Shutdown Event Tracker" and click the radio button to set it to "Disabled."


Windows NT

Can I install NT 4.0 on an NTFS partition larger than 4GB?
Yes, but not without jumping through some hoops. The problem is this: the maximum size for a FAT partition in Windows NT is 4GB. During the initial setup of Windows NT, NTFS partitions are not really created: they're converted. That is, NT first formats the partition as FAT, then later converts it to NTFS during the final stages of the installation. There are a couple workarounds. If you have another working NT machine, you can put the disk in that system, format it as NTFS, and then put it back in the new system. Or, as an alternative, you can partition the disk and install the system into a partition of 4GB or smaller during the installation; then, when the install is complete, use the Disk Administrator tool to format the rest of the disk and use the Extend Volume Set command to create a single volume which span the partitions.

How do I change how long it takes for NT to boot to the selected operating system?
Open the "System" item in the control panel. Click on the "Startup/Shutdown" tab. Set the value for "Show list for" to the desired number of seconds. Or, if you're a UNIX person who just hates all those GUI tools, you can also do this by editing the boot.ini file.

The explorer shell has crashed on my NT box. How can I get it back without restarting?
First, use the Task Manager to kill the current instance of explorer.exe. Then, from the Task Manager's "File" menu, choose "New Task (Run...)". Type "explorer.exe" in the "Open" field, and launch the program.

Can I change the drive letter assigned to my system drive?
Yes, but it's kinda painful, and you should ask yourself whether or not you really need to do it. First of all, it will involved editing the registry, and that is, as any Windows users knows, is a Task Which Can Destroy The Whole Damn House of Cards. Do you trust me to give you good advice on what to do? Remember, I'm a dumbass. Anyway, here's the procedure. In brief, you'll need to edit the registry to change all instances of the old mapping (let's say E:\) with the new letter (for our example, it will be D:\). First, you'll need to use the Disk Administrator to assign the desired letter to the system disk. You'll have to reboot to commit the change. Then, open regedit and search for the old mapping. Search keys, values, and the whatever else. When you find one, replace it with the new letter. If you haven't installed a bunch of programs, it will probably be less painful. And if you HAVE installed a bunch of programs, I'd seriously question whether or not it's worth doing. Remember, some programs may use .ini files or other such files to set path values, and they may not use the registry at all. If that's the case, you'd need to find all THOSE entries and change them, as well. So why would you EVER bother to do this? In my case, I needed to create a new server with a directory structure that was identical to a production server... and since I installed NT from a set of .cab files on a second hard disk, and that disk had a primary partition, the extended partition on which I installed NT wound up at E: instead of D:, and hey, who wants THAT? Anyway, if you do this, even on a system which is essentially blank, expect to spend about 20-30 minutes on this task.

Sometimes my password is case sensitive, and sometimes it isn't. What's going on?
Chances are you have more than one domain controller. At least one of your domain controllers is using Lan Manager compatibility, and at least one isn't. If you happen to log in to the server that supports LAN Manager, your password WON'T be case sensitive... but if you hit the other server, it will be. The default settings changed between SP3 and SP4, I think... in any case, there is an article on resetting this on Microsoft's support site:
KB Article Q147706.

I changed a user's password, and now some of my domain controllers aren't recognizing the new one yet. How can I force them to synchronize?
On the primary domain controller, open a command prompt and execute the command:

net accounts /sync

How can I extract the files from a Service Pack?
Assuming we're talking about Microsoft's Service Packs, such as those for Windows NT and Exchange, use the /x option from the command line. For example, if your Service Pack file is called sp6i386.exe, you'd type the following at the command prompt:

sp6i386.exe /x

You'll then be prompted for where you'd place the extracted files.

I'm getting an IP address conflict on my Windows network. How can I find which system is at fault?
Generally, the warning box that pops up in the event of an address conflict will contain the MAC address of the system which is conflicting. In Windows NT or Windows 2000, the event viewer will also show the conflicting MAC address. The problem is, how do you find which computer that MAC address belongs to? There are a couple methods. If the address in question has been handed out by a DHCP server, you can look in the DHCP Manager to find which computer has that address. If the address is no longer assigned to that computer, you can still possibly find which one is the guilty party by going to the properties of each computer in the DHCP leases. The properties will show the MAC address. (Keep in mind that you may not see the conflicting IP address in the list of DHCP leases. Conflicts can sometimes be caused by computers logging on to the network without releasing previously-held addresses; they might eventually get a different address from the DHCP server, but even a very brief conflict can cause communication problems.) If you're on a large network, you can use a free tool called NBTScan (available here on the DOS/Windows Software page) to quickly list all IP addresses and the corresponding MAC adapter.

One day I plugged a serial mouse into my NT box. Now that I'm using my PS2 mouse again, I always get an error message saying that a service or device failed to load at startup. How do I make it go away?
Go to the "Devices" item in the control panel. Locate the "Serial Mouse Driver" and set the startup to "Disabled." Reboot. That will solve the problem, but it will mean that a mouse would not be detected if you ever need to use a serial mouse again. To enable the serial mouse for future use, go into the Devices control panel again, and change the startup type back to "System."

I can't create new users! What happened, and what can I do?
Well, I've got a partial answer for you: you probably disabled some necessary services in the process of hardening the server's security. I did this on an NT4 webserver once, and although I can no longer remember exactly which service was the culprit (workstation, maybe?) I do remember how I got around it: although I couldn't create new users from scratch, as long as there is an existing account with correct permissions, I could duplicate it by selecting "Copy" from the User Manager's "User" menu. Presto... new user.

Why do shares sometimes not disappear in the server manager? Why am I getting these errors in the Event Viewer?
I'm not entirely certain about this, but I think if you delete the share itself, the server manager will know about it, and you won't get any errors... but if you delete a parent directory CONTAINING shares, server manager DOESN'T know. You've go to go the Server Manager and manually disconnect the share. Actually, I haven't been able to replicate this error on any of my current servers; I think it may have been fixed in Service Pack 4 or SP6.

I used a hard disk in a Linux system, and now I want to use it in a DOS system. I reformatted it, and I set the partition active, but I still can't boot from it. The LILO boot loader still keeps appearing. What's wrong, and how do I fix it?
You need to restore the master boot record. The boot sector of the disk is not affected by reformatting. Use this command to clear the boot sector:

fdisk /mbr

Once you've done this, assuming that your primary disk partition has been set active, you'll once again be able to boot to DOS. Oh, and a little side note: if you use the command "fdisk /?" to try to find out what you can do with FDISK, you'll be out of luck. Apparently, the folks at Microsoft didn't think it was important to document the procedure for fixing the boot sector.

I have a dual-boot Windows 98/Windows NT system. I had to re-install Windows 98, and now I've lost my boot menu. How do I get it back?
The problem of losing the boot menu can happen a couple different ways. I lost mine as a result of swapping video cards, oddly enough! Removing my ATI All-in-Wonder card somehow messed up my IDE controllers, and two of my hard disks were suddenly inaccessible. The only way I could figure out that was to re-run Windows setup; and doing THAT altered my boot sector, and left me with no boot menu. There are a couple different ways to fix this. The easiest is to use the Windows NT setup floppies to repair the boot sector. When you are prompted for what type of setup you want to run, choose "Repair a Damaged Installation." You'll then get a list of options for what you want to repair. Unselect everything EXCEPT "Boot repair." This will restore your boot sector.


Windows 2000

I can't find my "Send To" folder. Where is it?
If you don't see the "Send To" folder in the "Documents and Settings" directory, you will need to click on Tools in Windows Explorer/Folder Options/View and check "Show Hidden Files and Folders."


Windows XP

How can I open the Device Manager from the command line?
You can get to the Device Manager by wading through a few clicks from the computer properties screen, but there is a faster way to get there: From a command prompt (or from the "Start | Run" dialog), just type the command:


I'm connecting to a remote machine using RDP, and I want to access my local hard drive. In the "Local Resources" section of RDP, I've configured my machine to allow access to my hard drive. However, when I log in to the remote machine, I don't see my drive. What's wrong?
I've encountered this problem when connecting to the console session. My first reaction was that the console session doesn't "know" that you are using RDP, and hence it doesn't know that you have available resources for sharing. (If you're wondering how to connect to the console session -- or if you are connecting to the console -- check the properties of the shortcut you used to launch RDP. Prior to XP SP3, the parameter for the console was "/console." For machines using SP3 or later, the parameter is "/admin.") Later, however, I discovered that it isn't quite as straightforward as that: sometimes I can see the local drive when I'm connected to console. My latest theory is that you can't see shared local drives unless you have previously connected via a regular RDP session with drive sharing enabled. Once you've done this, you can then disconnect and reconnect as console, and your shared local drives will be available.


Microsoft Exchange

I'm trying to create a mailbox in Exchange 5.5. When I try to assign the email address I want, I get a message that "This email address already exists in this organization." I checked all my users and no one has that address. What's going on?
You probably have a public folder with same name as the desired address.

We're using Outlook Web Access with Exchange 2003. Some of our users have a question mark (?) appended to the beginning of the body of every message they send. What's going on, and how do I fix it?
Outlook doesn't know how to deal with the MIME type for your message. Try this: open Outlook Web Access, and in the lower portion of the left-hand pane, click the "Options" tab. Scroll down to the section on "Email Security." Hit the "Download" button where it says, "Click here to install the latest version of the S/MIME Control. When the file download dialog appears, choose Open."

Where can I find the settings to limit message size in Exchange 2003?
Although these settings can be overridden by local values for particular users, the default maximum sizes in Exchange 2003 is 10240KB (i.e., 10MB). These settings are configured in the Global Settings/Message Delivery properties in the Exchange System Manager.

Here's a fun fact, though: my own experience suggests that these limits are imprecise. Attachments have to be MIME-encoded to pass through email, and the size of the encoded file is not the same as the original file. Sometimes it will be larger, and you might find messages being rejected due to excessive size, even though they appear to be below the limit.

How can I open another user's mailbox or calendar using Outlook Web Access?
First, let's state the obvious: this only works if your account has been given the appropriate permissions to allow you to access that folder. In the regular Outlook client, you'd be able to access the folder by going to File | Open | Other User's Folder... and selecting the folder. In OWA, however, there's no such menu. However, you can still access the folder with a little workaround.

Suppose I'm Tom Jefferson, and I need to access Abe Lincoln's mailbox. My mailbox login is tjefferson. So, I log in to OWA as tjefferson. If I look at the address bar when I first log in to my mailbox, I might see only:

That doesn't tell me much. However, if I open one of my messages, I see:

The first part of that string shows me the URL for my inbox. Now that I know the URL, I can close the open message, and I'm back at my Inbox. Once I'm there -- without closing the browser -- I can paste the following URL into the address bar:

If I have the appropriate permissions, the mailbox should open.


VMWare, Hyper-V, Virtual Server, and Virtual PC

How can I mount a VMWare disk file so that it appears as a hard drive on the host machine?
To do this, you'll need the vmware-mount utility. This can be downloaded from the VMWare site. (I believe it is now included in the VMWare Server download, as well.) Suppose your VMWare disk file is in c:\vm_folder\myvhd.vmdk, and you would like it to appear on the host as drive J. To mount the disk, use the command:

vmware-mount j: c:\vm_folder\myvhd.vmdk

And to unmount the disk, use either:

vmware-mount /f j:
vmware-mount /d j:

I converted a virtual machine from Virtual PC to VMWare, and now my mouse isn't working correctly. How can I fix it?
This bug can occur if you've installed the Virtual PC (or Virtual Server) additions before installing the VMWare Tools. After installing the VMWare Tools, your mouse cursor will bury itself in the lower right corner of the screen, and will refuse to move. A discussion of this problem (and a few different solutions) can be found at The best solution (i.e., the simplest) is to edit the registry in the virtual machine. Browse to the key HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Contro l\Class\{4D36E96F-E325-11CE-BFC1-08002BE10318} and remove msvmmouf from the UpperFilters Regvalue. The reboot the VM.

My Hyper-V server is unable to manage its virtual machines, and displays a message saying, "The operation on computer 'localhost' failed." However, all of my virtual machines are still working. How can I fix the problem?
I can't explain precisely what causes this problem, but it's generally easy to fix. One solution is to reboot the Hyper-V server. Obviously, if you've got a bunch of production VMs running on the host, that might not be such an easy fix. Alternatively, you can restart Windows Management Instrumentation from the Services control panel. When you restart WMI, several other dependent services will also automatically be restarted. Once they've been restarted, you should once again be able to administer your machines via the Hyper-V Manager.


Visual SourceSafe

I'm trying to create a user in Visual SourceSafe, version 6, and I'm getting an error that says "Invalid DOS path." What is going on?
The program is missing an .ini file, specifically template.ini. Why is it missing? Who the hell knows. But Microsoft does have a Knowledge Base article (149378) that explains how to fix it: you create the file, and drop it in the Users directory on your SourceSafe server. Here's what is in the file:

   ; ss.ini
   ; This file contains all the variables that "customize" Visual
   ; SourceSafe to your particular needs. The ss.ini variables are
   ; documented in the Visual SourceSafe User's Manual. Only a few of them
   ; are placed in this file by default.

   ; Visual C++ programmers should remove the semicolon from the following
   ; line, to uncomment it.  Other programmers REPLACE the line with
   ; different masks.
   ; Relevant_Masks = *.c, *.h, *., *.asm

   ; The following line prevents you from being asked for a checkout
   ; comment.
   Checkout_Comment = -

   ; The following lines force Visual SourceSafe not to execute certain
   ; file types.
   .reg (Win) = notepad.exe
   .vbp (Win) = notepad.exe
   .vcp (Win) = notepad.exe
   .mak (Win) = notepad.exe
   .bat (Win) = notepad.exe

   ; Your current Visual SourceSafe project.
   Project = $/

For other "fun sourcesafe errors" you might also want to take a look at Microsoft's KB article 176909 on moving a SourceSafe database. This article also concerns missing files... imagine that.

More tips coming eventually...