I was having problems removing Silverlight 3 from the Add/Remove Programs feature. These are the steps I followed to get rid of it and upgrade to Silverlight 4.

Step 1 Create a new system restore point (if you feel like it)

Click Start, All Programs, Accessories, System Tools, and then click System Restore. The System Restore Wizard opens. Click Create a restore point, and then click Next. In Restore point description box, type a description. Note:  The date and time are automatically added to your restore point. Therefore, you dont need them in your description. Click the Create button.

Step 2 Try Add/Remove Programs Again

In the control panel find Add/Remove Programs or Programs and Features whichever one you have. Uninstall Microsoft Silverlight if it is shown as installed by clicking on it and then clicking Uninstall or Remove.

If you still get an error message (for example a dialog appears saying that the msi file cannot be found) just continue on to the next step.

Step 3 Mess with things

Click on Start > Run  Type in cmd and select OK. From the command prompt screen run the following commands

cd\
reg delete HKLM\Software\Microsoft\Silverlight /f

Close the command prompt window.

Rename the C:\Programs Files\Microsoft Silverlight folder to C:\Programs Files\MSSL

Goto Start >Run and type Regedit. It will open registry editor. Highlight My Computer. Click on the File menu and select Export. Make up a file name and save the file (on the desktop for example).

In the registry editor navigate to HKEY_CLASSES_ROOT\AgControl.AgControl Expand AgControl.AgControl

Select the CLSID key. Edit the (Default) value by right clicking and selecting modify and change it to {DFEAF541-F3E1-4c24-ACAC-99C30715084A} then click OK

Select the CurVer Key and change the (Default) value to AgControl.AgControl.1.0

Navigate to HKEY_CLASSES_ROOT\Installer\Products  Find the key called D7314F9862C648A4DB8BE2A5B47BE100 and rename it to OLD_D7314F9862C648A4DB8BE2A5B47BE100

 
Step 4 Try to install Silverlight again

When I hear about Exchange Servers and Active Directory, I used to get irritated because I knew little about it. So, I decided to read up and write up. This is not an administration guide, just an overview, so sit back, relax and scroll down.

What’s Active Directory?

To me “Active Directory” is a dumb name, it doesn’t mean anything. Active Directory is meant to be a Directory Service. There are many types of Directory Services, for example a phone book or directory enquiries. They contain information (in a phone book, the information is about phone numbers) and it’s a service that you can use (eg a book that you can look through, a phone number you can call or a website you can visit).

Active Directory is the directory service included with Windows servers. It stores all the information you would need to manage a windows network, for example user accounts, file servers and printers, and the rules about who’s allowed to do what (Group Policy Objects or GPOs).

You might have noticed the word Object pop up there in GPO. The data stored in this ‘Active Directory’ is organised into things called objects. The Active Directory Schema is where you define what kinds of objects you can put in your directory. If you’ve never heard of objects and schema and you are a bit confused, then I suggest you google it, it’s quite similar to XML schema.

It’s a confusing word “object”, because it obviously means something special in this context.. but it’s an abstract term, so it’s hard to remember. In my head, I substitute object with “thing”, because I can understand things. I don’t understand objects because “object” weird

Lots More Confusing Words

I think Active Directory should actually be called ‘Lots of Confusing Words’ because that’s what it is, and that’s what’s coming up next.

In Active Directory you have to organise all your information into objects, so you need to put it into some kind of structure. So Microsoft made up all these things to help you structure your directory, and here they are.

All your computers are probably networked. So you normally organise your LAN into a Site. Site means your physical site. You group things into sites because you need to distribute directory information across your whole windows network, so organising your computers into sites helps you manage this information sharing.

You put all your users, printers, email addresses, databases etc. into domains. You can make one domain and use it over different sites or locations. So if your organisation has two main sites in two cities, you don’t need to make two domains, you can use one big domain.

Domain Controller – Oh yes, the DC, something you can touch. This is your active directory server that keeps a copy of the domain directory. You can have more than one domain controller in a domain. A domain controller can only service one domain. A domain controller is the server that checks user logins etc.

Trees – this is when you create a domain hierarchy. By making a tree of domains in your active directory, you can administrate and manage your network more easily. This is actually quite a simple idea, and it’s tied in to DNS a bit. Imagine you have a domain name cyleft.wordpress.com. wordpress.com is the top level domain or main site, and cyleft.wordpress.com is my bit. There are lots of ‘child’ domains under wordpress.com for everybody else who uses wordpress like torinelson.wordpress.com, everyone looks after their bit. So you can think of it like a tree, with wordpress.com at the root, and cyleft, and torinelson as two branches coming out from there, among others. That’s what your tree is, a domain tree helps you organise your directory.

Trees are fine, but Forests? Are you kiddin? If for example, you had blogspot.com as a domain tree and wordpress.com as another domain tree, you can make a forest! Whoopeee! Now you can share Active Directory Schema, and communicate between trees etc. Both trees trust each other, so if you logon to one tree, the other tree trusts that you’re authenticated and logged on (otherwise you’d need to login to both trees)

Organisational Unit (OU) is another confusing name. You make OUs to help you administrate your network, so you can have lots of administrative groups to manage your actual windows network. You can put OUs in other OUs to make a nice tree as well.

If you have more than one domain in a forest, you need a global catalog so that you can login no matter which domain controller you are using. That’s what the global catalog does, it stores all group membership information. This is usually stored on a global catalog server and cached on each domain controller.

Users and Groups

Now when you login to your computer, you can login in two ways. One, your username and password is stored and checked on your computer, known as a Local User Account. The other place you can store and check your password in is the Domain Directory on a domain controller, making it a domain user account. This means that you can have local administrative accounts that can control your computer, and domain administrator accounts which let you change things in the domain, tree or forest.

You also have user groups. These are just to group users and things together. You can have different types of groups, some are limited to a single domain, and some groups let you grant permissions and access to anything in any domain in the forest.

Here comes Group Policy. It should really be called Grouped Policies I think. A GP is a collection of settings that tell you how programs, the operating system etc. work for users and computers in the active directory. You save them in a group policy object and apply it to domains, sites, OUs etc. It doesn’t mean that you create some policies or settings that you apply to a group, even though that’s how it sounds. GPs are not applied to groups, you apply them to domains, or OUs or whatever. Stupid I know. You can have local GPOs for your local computer and domain policy GPOs which get applied to every user in the domain.

Tools & Utilities

I’m just going to put in a list of tools and utils used to administer/diagnose active directory. If you want to know more, leave a comment:

• netdiag.exe
• dcdiag.exe
• ntdsutil.exe (NT Directory Service was the old name for Active Directory type stuff in years gone by)
• dcpromo
• gpedit.msc
• netdom
• dsastat
• repadmin
• dsadd, dsmod, dsquery
• whoami
• runas
• gpresult

There are lots of details about Active Directory which will bore the hair off your eyebrows but I have already had enough of Active Directory now, so I will conclude here. The majority of Active Directory is straightforward, there are just lots of new words to hate.

Start Menu -> All Programs -> Accessories -> System Tools -> Disk Cleanup

If Disk Cleanup is taking an annoyingly long time to start up because it is ‘scanning for old files to compress’, you can force it to stop this scan. All you need to do is modify the registry. Using regedit.exe or similar tool, browse to this key:

HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows\CurrentVersion\explorer\VolumeCaches

Find the ‘Compress old files‘ registry key and rename it. Save and close.

You can find Microsofts knowledge base article on this here:

This post was prompted by a comment on one of the articles, so I’ve moved it to it’s own spot to write a little about it

I want to find a totally free hosting service that is ok to accept CGI. Do you believe that I can have a wonderful host or better a first-class free hosters database or something like that for what I want ?? ps: Do you know free no ads hosting ?? How have you built your blog ? — Quentin Bonnes

OK, you can have wonderful host, a fantastic host, even better, a first class, top notch, top of the line, triple-a host. But I don’t think you can get one for free.

Do I know any ad-free, cost-free hosting service? Nope, most have ads.

How have I built my blog? Why, that’s easy. All you do is go to www.wordpress.com and hit the big orange button that says sign up. It’s a quick and easy and free. The editor you get to write your blog with, is amazing – very easy to use.

If you’re looking for hosting with features like CGI script support, databases etc. I’d go for a ‘cheap host’. They’re usually around £3-5 a month. The web servers might be a bit slow, but they’re good value for money.

If you’ve found a web host you like (or not) get in touch!

Solution: CMOS battery. PC would boot if the battery was removed. If it is inserted, system doesn’t power on. Replaced battery, but the problem persists. Perhaps there is corrosion somewhere near the battery socket which causes this. I just left the battery out.

“When trying to update a password, this return status indicates that the value provided as the current password is not correct.”

SOLUTION: Replace SECURITY/SAM file in C:\WINDOWS\system32\config

Background

Cloning a hard drive means making an exact copy of a disks contents so that you can either backup or migrate your disk. This post is about cloning a windows partition so that you can run it from a different PC with different hardware on it. There is a useful section at the end, which will help you get into windows if you’ve forgotten to sysprep.

Overview

An easy piece of software to use is Ultrasoft Snapshot. Then use Microsoft sysprep utility to prepare the system for cloning. Use ‘snapshot’ to clone the partition to a backup hard drive. Restore the backup to a new disk.

If you haven’t run sysprep, and you no longer have access to the old machine, there’s some help to get you going in the trouble shooting section.

Microsoft sysprep

Microsoft have a page on sysprep. I forgot to use sysprep (hence this post) so you’ll have to google this section. Good Luck.

Ultrasoft Drive Snapshot

Use the ‘snapshot’ software from a startup disk in your original machine (eg bartpe) or by installing the original hard drive in another pc (let’s call it the host pc) and running ‘snapshot’ on the host machine. This is required because snapshot needs exclusive access to the source hard drive partition (the one you’re cloning), so you can’t be running windows off the drive you’re cloning whilst cloning.

You need the same amount of space free on your target hard drive partition as the source partition. If you have alot of free space on the source partition, it might help to resize it, eg using Partition Magic.

Use the ‘snapshot’ program to extract the cloned partition to the target hard drive. You can then boot into your target machine and finish. If you have trouble booting, mark the partition as active (can be done from boot cds, or installing drive in different machine and using the computer manager from the windows control panel.) Make sure the MBR has a working boot loader (eg use fixmbr from windows installation cd restore mode to reinstall bootloader.)

Troubleshooting ie forgetting to sysprep

If you cloned the partition and tried to boot it without sysprep’ing before cloning you will probably get problems starting the computer. It either restarts endlessly, or you get a blue screen (STOP 0x0000007B messages indicating problem with boot drive).

This usually happens because windows has problems starting in the new machine, as it is still setup for the old machine. One way to solve this problem is to install windows on your target machine in a new & empty partition. Then you need to copy over the system/sam/ security files from C:\windows\system32\config over to the problematic install. This will effectively overwrite the data windows keeps about the system hardware. I already had a windows installation on my target machine, so I copied these files over, and in my case this allowed me to boot into the cloned windows partition!

The only problem then remaining is that you might have registry entries in your old config files that are required by the software you had installed. These have to be manually sorted out by mounting the hives in regedit and copying the required data out. This is possible for a few registry entries (eg Windows Services entries etc.) but it’s probably too much hassle otherwise.

TODO

My problems were solved after copying over the configuration files, and fixing a few registry entries.

If you have time, instead of overwriting the entire SAM/SECURITY/SYSTEM hives, only overwrite the sections containing hardware information. Which sections are those? No idea, you’ll need to find out. Copy out enough registry entries to get it to boot (may only be those regarding chipset/graphics card/hd.)

Otherwise, find a guide on sysprep’ing after cloning and try that. Oh and let me know if that works.

To use, you’ll need to download and install MPLAYER as well as the extra plugins. You may also need Real Player, Quicktime, Windows Media Player installed (or an alternative.)

OK I’m always having to lookup the mencoder flags for encoding video files, so heres a useful example:

c:\mplayer\>mencoder.exe input.file -of mpeg -ovc lavc -oac lavc
                  -lavcopts vcodec=mpeg2video:acodec:mp2 -o output.file

of = Output File ovc = Output Video Codec

The rest of the options can be looked up.

Start -> Run -> regedit
Edit -> Find

In the search box type this:

{87D62D94-71B3-4b9a-9489-5FE6850DC73E}

Tick the box titled ‘Keys’ and untick the others. It should find the key for ‘AVI Properties handler.’ Right click the key from the left window pane (with the folder tree view) and click export. Choose a file name to export the key to, and click OK. Now you can delete the key you just exported. If anything goes wrong, you can double click the .reg file you just created to restore the key.

You can also delete the following keys:

{40C3D757-D6E4-4b49-BB41-0E5BBEA28817}
Video Media Properties Handler

{c5a40261-cd64-4ccf-84cb-c394da41d590}
Video thumbnail extractor

The files I had originated from a Microsoft SQL Desktop Edition 2000 (MSDE) server. I’m assuming these instructions will work for other versions as well.

Download MSDE2000

You can download the 42MB MSDE 2000 Release A file from here:

Install MSDE2000

If you run the file you saved it will ask you to choose a folder to extract the setup files to. Please select a folder, or just click next. The files will unpack.

Open a command prompt from Start ->All Programs -> Accessories -> Command Prompt

It will show you something like this:

Microsoft Windows XP [Version 5.1]
(C) Copyright 1985-2001 Microsoft Corp.

C:\Documents and Settings\username>

Type the following in and press enter:

C:\Documents and Settings\username> cd C:\the\folder\you\extracted\your\files\to\goes\here
C:\the\folder\i\extracted\my\files\to> setup.exe SAPWD=YOUR_PASSWORD_GOES_HERE

This is the type of window that should show next

Setup will commence, and you should have your server installed.

Graphical Administration Tool

I tried this, but couldn’t get it to work. I wasn’t using IIS. Good luck.

Command Line Administration using OSQL

OK back to basics. osql.exe is in this folder:

%ProgramFiles%\Microsoft SQL Server\80\Tools\Binn\

By default this is:

C:\Program Files\Microsoft SQL Server\80\Tools\Binn\

Open a command prompt and cd into that directory:

C:\SQL Server\80\Tools\Binn\>osql /?
usage: osql              [-U login id]          [-P password]
etc

Read the stuff that it prints out here.

The -E’ switch means trusted connection, so we’ll use that to login

C:\SQL Server\80\Tools\Binn\>osql -E

1> sp_help
1> go

A whole load of text will fly by. This is the output of the sp_help command. As you see, you use ‘go’ to execute the commands you type in at the prompt.

Attaching the MDF/LDF Files:

To view your MDF/LDF files  you need to attach the database files to the server. The commands are as follows:

1>EXEC sp_attach_db @dbname = N'CHOOSE_DATABASE_NAME_AND_TYPE_IT_HERE',
1>   @filename1 = N'C:\YOUR_FILE.mdf',
1>   @filename2 = N'C:\YOUR_FILE.ldf'
1>go

To view info on the databases currently attached, you can use these commands:

1>select * from sysdatabases
1>go
1>select * from INFORMATION_SCHEMA.TABLES
1>go

You can now use SQL statements as follows:

1>USE YOUR_DATABASE_NAME
1>select * from YOUR_TABLE_NAME
1>go

If you want to output the result of your SQL query in a file use:

C:\...\Binn>osql /E /d YOUR_DB_NAME /Q "select * from YOUR_TABLE_NAME" -o outputfile.txt

If you want to do it using a host/username/password to login:

C:\...\Binn>osql /U sa /P YOUR_PASSWORD /d YOUR_DB_NAME /S your.server.name.or.ip.address /Q "select * from YOUR_TABLE" -o outputfile.txt

Some more useful commands:

1>sp_helpdb mydb
1>go

1>EXEC sp_grantlogin ‘Test’
1>EXEC sp_grantdbaccess ‘Test’, ‘database_name’
1>EXEC sp_addlogin ‘test’,'hello’
1>EXEC sp_password ‘OLDPASS’, ‘NEWPASS’,'Test’

Have fun.