Mr Snrub

So you can start the machine and log on without a problem, the task bar & start menu are presented, but no wallpaper and no desktop icons (not even Recycle Bin) appear? If you right-click on the blank desktop, what options are presented on the context menu? If you click the Start button then click your user name, does it open an Explorer window with the various folders in your profile, one of which should be "Desktop"? If so, if you drill down into Desktop do you see your icons in there?

The program itself has to support the passing of parameters on the command line in order to use the simplest type of association.As an example, Notepad is associated by default with .txt files, so entering "notepad foo.txt" at a command prompt would attempt to open foo.txt, and promt to create it if not found. To see if gallery.exe accepts command line parameters, open a command prompt and navigate to the folder where it lives (using "cd"), then enter a dummy command line: gallery foo.gal If nothing happens at all, does a process start if you just run gallery.exe by itself? If the program launches but reports no error, then I would guess it doesn't check for any command line options and was written to be interactive only. If the program reports an error that it cannot find or open foo.gal, then file associations should be working...

The 3 messages report "specified in the registry", so you should be able to run regedit and search for a part that should be unique - I would go for asdlfj3. Also, I would guess that this is a single registry entry that is not surrounded by quotation marks (so gets treated as separate strings) - if you piece together the 3 strings you get: C:\User\name\Local Settings\Application Data\asdlfj3\services.exe Finally, you state that logging on as another user does not report the errors, so I would suspect the rogue registry value lives under an HKEY_USERS key (mapped as HKEY_CURRENT_USER when logged on as that user). I guess the value ought to be surrounded by quotation marks and/or this is dodgy - the only services.exe I know of is the Service Control Manager (SCM) which is part of the OS and not stored in individual user profiles...

Is the IE crash reproducible 100% by visiting certain URLs? Check in Tools / Manage Add-ons - what is listed there that isn't Microsoft Corporation and is enabled? Do you run AV with some kind of "web guard" agent?

Frequently Asked Questions: Windows Server 2008 Service Pack 2 and Windows Vista Service Pack 2 Windows Vista Service Pack 2 Deployment Guide

It's not just DLL dependencies that could cause issues, but the user profile folder structure and handling of junction points, hard links, etc. is very different - then there may be deprecated APIs due to the absence of "classic mode". (Those who have used roaming profiles from NT4, W2K and Vista days will likely be familiar with the changes to profile structures.)

To enable:bcdedit /set {current} usefirmwarepcisettings ON To disable: bcdedit /set {current} usefirmwarepcisettings OFF To remove entirely: bcdedit /deletevalue {current} usefirmwarepcisettings (I did these in a VM running Win7 build 7134 to see if I could repro your issue.)

Have you ever tried Vista on this laptop, and if so did it do the same? Does the same occur whether you start from powered off or restart? Does the system sleep and resume successfully? The value for timeout has no bearing on this if the countdown timer isn't being displayed at all, so you can ignore that. Can you paste the output of the 2 commands: bcdedit /enum /v reg query hklm\bcd00000000\objects (HKLM\BCD00000000 is where the BCD is mounted as a hive during boot - it's probably not wise to go writing in there manually ) Regarding the BIOS - are there any updates for it, and what settings do you have related to the disk controller and its I/O mode? I have only seen this symptom on a system with RAID1 where the BCD did not have all the resumeobjects set correctly, so if the first disk was pulled then the system would sit at the boot menu indefinitely until a key was pressed.

I use the built-in Windows Firewall as I have grown tired of adding exceptions for programs to make outbound connections and maintaining the list on a host firewall. It has little to do, however, as the NAT provided by my router prevents external clients connecting anywhere except my web server. For AV I don't like to trust to free software, so I have been using Avira on Vista x64 and Windows 7 x64 without complaint. And Windows Defender I let "do its thing" for other spyware/malware detection - as I set the PC to Sleep rather than shutdown, I rarely have to check for on-startup or on-logon items, but I check those with AutoRuns periodically.

The boot menu appears if the default GUID is not found - in this case the BCDEdit output would have the "Windows Boot Manager" value "default" listed as a GUID instead of "{current}". If this is the case, enter the following at an elevated command prompt: bcdedit /default {current}

From an administrative point of view, you would need to add exceptions in 2 places (programs and/or ports). Functionally, one of the firewalls has to be last in the chain - it may appear to receive all outbound packets from a single source (the other firewall) and so once this source is "trusted" it can render the last one obsolete. As with AV products, firewalls employ kernel and user mode components - the kernel filter drivers can sometimes "fight" to be layered last so that they have the last word on outbound packets and first word on those inbound (or a later driver could potentially modify the data after/before inspection). Even with no conflicts, every network I/O has to travel through both filter drivers and if SPI is used this might have a significant impact on the throughput (given network I/O is considerably slower than local I/O). 1 perimeter firewall (e.g. built into a router) and 1 host ("software") firewall per machine should provide a more diverse, layered approach. I've not seen problems with 2 host firewalls present at the same time, but then I've never tried I have, however, seen the havoc 2 realtime AV products can cause (hard hang after <5 minutes uptime).

bcdedit.exe is a command-line "administrative task" from Windows' perspective (if it was a GUI app it would have the shield symbol superimposed over its icon to indicate this).A member of the Administrators group is still subject to User Account Control - it can be argued that these users are more appropriate for UAC due to the greater risk if something malicious ran in their context. (The one and only exception to UAC when it is enabled is the Administrator account, it has a static, well-known SID and the account is disabled by default.) You have 2 security tokens when logged on as an administrator - one user and one administrative. By default, processes are started with your user token, so if the process attempts to perform an administrative task it will fail with "access denied". (The programs can have a "manifest" provided to change this behaviour, but for cmd.exe it is the default.) If you right-click the Command Prompt shortcut in Start / All Programs / Accessories and click "Run as administrator" then it will launch the process with your administrative token - this will trigger a UAC popup to confirm you really want to do this - then administrative programs (such as bcdedit.exe) can execute happily. Basically, the whole concept of UAC to is alert the user when their account wants to start a process that has the potential to change part of the system itself - by having the user "sign off" these alerts, they are at least made aware when something may have a greater impact than they realise, or if a process is being launched without their knowledge (malware/spyware).

You had outstanding updates - see the warning symbol on the Shut Down button? If Windows Update detects there are applicable updates for your system, the default action there is to "install and shut down" (I think the tooltip even says this if you hover over it). (If you run a check for updates and install them, the button returns to its normal setting.) Also, if you have installed an update that requires a restart you get a similar thing I think.

Nope. Windows XP Pro x86 and x64 use different keys, not interchangeable. Same story for Windows Server 2003 - Vista was the first version to have a "SKU-based" product key that works for any platform.

ALT+[up arrow] can be used to "go to parent folder"

Did you see the edit I made to my first reply, and did you do the diagnostics testing cluberti recommended twice? http://www.msfn.org/board/Random-and-many-...st&p=795567 http://www.msfn.org/board/Random-and-many-...st&p=795666 http://www.msfn.org/board/Random-and-many-...st&p=795793 I'll start downloading the dump when I get home and get back to you.

The key combination is for producing a manual dump, usually for when the system is hung, you don't need to do that. When the bugcheck occurs the memory dump will be produced - you just need to wait (or provoke the bugcheck) and then get C:\WINDOWS\MEMORY.DMP zipped & uploaded.

OK, no problem, I guess CRASH is what we are looking for (more suitable to the case) but just one question:what is crashing proglem we are monitoring? what should I write in MYAPP.EXE: adplus -crash -ctcf -pn MYAPP.EXE -quiet -o C:\AppDumps Nope, you are experiencing a bugcheck (blue screen) so you want the section entitled "KERNEL-MODE MEMORY DUMPS" - this is what makes Windows produce a memory dump of the whole system when it crashes.

You could also try disabling the network adapters in Device Manager to see if it makes any difference. So nothing changed on the system just before the problem started? No new software installed or upgraded? Not connected to a new network? Started immediately after a Windows Update? Follow the steps here for configuring the system to create a complete memory dump, then compress & upload it somewhere for analysis.

Okay, more shooting frmo the hip then Check if there are any BIOS, device/chipset/display driver updates for the laptop. Was it shipped with Vista pre-installed and had this problem since day 1, or did something get changed afterwards? Does the problem occur if all network adapters are disabled in the BIOS? (wired and wireless)

You could try installing the hotfix here, to see if it helps: http://support.microsoft.com/default.aspx?...kb;EN-US;952681 I assume you're not hot-plugging any PCI-E device manually, but I wonder if you have a PCI-E graphics card that maybe occasionally isn't getting enough power so it "drops off" and could be considered a hot unplug? (Just a theory.) Edit: Also, could it be overheating? Does the problem tend to occur after the system has been on a while? Check the grilles are not blocked or clogged up with dust, I've had someone with bizarres issues including poor performance when the fan grille was blocked and the CPU stepped down its speed due to overheating.

If it's static content, it's as simple as copying the files and creating the new site on the W2K8 server. If it's dynamic content using databases then you may have a bit more of an interesting time setting up the ODBC or OLEDB connections (I know it's the bit that I hate when I migrate my site to new hardware and why I put mine into a VM!). Avoid using FQDNs or explicit paths (use virtual paths) and migration becomes trivial.

First (and possibly only required) step is to create an A record in DNS for www.intranet.com and point it to the IPv4 address bound to the website in IIS. If you have multiple websites on the same IP address in IIS then look into "host headers" and add the name there, so IIS knows which site content to deliver when a request arrives.

There is no redistributable package for SP1, as it was the release version of XP x64. Service Pack 2 can be downloaded here, however.

Perfectly normal - same kind of setup I have here.The cable modem forwards the DHCP requests to acquire public IP addresses, so clients that are connected to a SWITCH will get unique public IPs (I have up to 5 through my ISP). When a ROUTER is used, it will obtain 1 public IP address for its WAN interface, and act as a DHCP server for the internal clients who will get unique private IP addresses - from the Internet these clients would all appear to be behind the same public IP address. When the cable modem is connected to a LAN port on the router, the routing logic is not being used - the device is being used as a switch, same as in the first scenario above. My network as it stands today: The clients and server are all behind the DIR-655 router, so only 1 of my 5 public IP addresses is used. If I needed to use a different public IP address, I can connect a client to the DES-1005D switch - but I don't like to connect clients or server to dirty networks. There isn't really much that requires clients to have unique public IP addresses, if a UPnP router is used.