Wherever APIs are mentioned the JSON format I not far away. Since I’m using two tools regularly I would like to introduce you to them.

JSON Viewer

If you only see the JSON-Text you are usually not able to see the structure. With the help of JSON Viewer you can have an easy overview: [URL]

JSON2CSharp

The name says everything: with this tool you can easily transform JSON into CSharp code. That’s pretty useful if you are, for example building the opponent to a JSON-API and want to work with the results. [URL]

Although the Metro Design focuses a lot on Typography Icons are still quite important. If you install Windows Phone SDK you will receive 36 Icons. You can find them here:

C:\Program Files (x86)\Microsoft SDKs\Windows Phone\v8.0\Icons

Unfortunately many Icons you might know from the common applications are not integrated. Pedro Lamas extracted 99 additional Icons from the App-Resources and offers them as a download.

I was working with ADFS 2.0 (“Active Directory Federation Services”) for a while when this simple question crossed my mind: How can I figure out if the connection between ADFS and AD “works”? Here is a simple test…

What is ADFS?

If you need some “position of trusts” beneath the AD-boarders you choose an Active Directory Service in the world of Microsoft. They are communicating between the dispatcher (own company-AD) and receiver (another AD or a big “center” like for example the Windows Azure Access Control Service) and issues claims for the registered user. Maybe this isn’t 100% accurate and maybe my choice of words doesn’t fit 100% but that’s how I understand the system

So, how do I test the functionality of the ADFS?

The ADFS uses the IIS to host his own end points. There is also a simple Login-page that every user can use:

https://{ADFS-FQDN}/adfs/ls/IdpInitiatedSignon.aspx

Afterwards a simple „Login-Page“ appears – after one click on „login“ you should see something:

If this site appears without username/Password the login works over Kerberos – otherwise you should use NTLM.

If everything goes wrong (or the configuration database is “broken”) you will receive an error message like this:

What you can test with it

With that you just make sure that the configuration/connection between ADFS and your own AD “works” – not more – but it is possible that the problems appear already at this point. If the “opposite site” woks or not is another question.

I have an ADFS proxy running – what’s next?

Basically you test the “main” ADFS first and later you have a look from a different machine which is only looking at the Proxy to make sure the Login works. Afterwards it goes on until the “customer”.

In this blogpost you will find a better description (and I think that’s also where I found the advice): How to test if ADFS in functioning

Troubleshooting

I had some problems with the ADFS which are not totally solved but I still found some links which might be helpful for someone:

AD FS 2.0: How to Change the Local Authentication Type

AD FS 2.0: How to Configure the SPN (servicePrincipalName) for the Service Account

Better blog.mydomain.com or mydomain.com/blog? Good question! If got asked this question again via Twitter on the weekend so therefore I decide to share my experiences:

Choose a subdomain, if….

- You plan to offer “different services” which are “logical separated” on one domain

- You are able to influence the subdomains without much effort

Example:

Blog.mydomain.com

Shop.mydomain.com

Forum.mydomain.com

Why?

If you choose to use subdirectories it is much more difficult to age these services. It is easier to create a CNAME entry and with that for example blog.mydomain.com leads to Tumblr Blog and not to a machine. Also it is possible to run the different sites on several machines (without Loadbalancer and so on).

Choose a subdirectory, if….

- You aren’t able to influent the subdomains (for example the whole applications-stack is installed at the customer but you don’t want the admin to unnecessary implement DNS-Settings

- The Application-Intern or the site is too small to create an own WebApp (like an Impressum for example)

What does Google think about this?

Here is a statement from a Google employee on YouTube.

And what about Stackoverflow?

I really like this answer:

Pro’s for subdomains:

· You can isolate configuration (for for example apache) per-domain.

· It will be easier to migrate parts of your application to other machines. Sub-directories won’t really give you this flexibility.

· Instead of having to use a $baseUri variable in every html template, you can just assume the root of the app is always /.

Cons:

· It will be much more annoying to quickly setup staging or temporary development environments. For every ‘app’ you will now need DNS of hosts-file entries and webserver configuration. With subdirectories you could drop the app in a directory, and go!

· If you do ever have the requirement to deploy your application on a different system where using / is because of some odd policy not possible, some rewriting might be in order.

My advice:

Make sure you can always do both, which will give you the best of both worlds. Every part of your app should have a configurable base uri that is always respected. As long as you make sure you can always go both ways, then who cares what you do? it’s just a url and it can always be changed.

But aren’t you using code-inside.de/blog yourself?

Well – when I created this blog some years ago I just didn’t know it better. The problem with a change would be that the old Links still have to work and I’m not sure how Google would take the news.

Therefore: think about this before you start

Download-Link or MSDN Link

I found the announcement today on the blog of Vittorio Bertocci.

More information’s?

If you want to get deeper into the subject you should risk a look on Vittorios Blog. You may be also interested in his Video about Windows Azure Active Directory.

From SAML to JWT

What is not included in the PDF but might become a trend anyway is the renunciation from SAML and to JWT. I recommend Vittorios Blog again and the Windows Azure Authentication Library.

Unfortunately there are not so many sources for information’s about this at the moment so you won’t be able to avoid his blog anyway

Azure Website Configuration

At the adjustments of Azure Websites you will find three diagnostic-tools:

· Detailed Error Logging – Turn on detailed error logging to capture all errors generated by your web site.

· Failed Request Tracing – Turn on failed request tracing to capture information for failed client requests.

· Web Server Logging – Turn on Web Server logging to save web site logs using the W3C extended log file format.

Source: How to monitor web sites

Where are these Logs saved?

The logs are saved in a directory on the machine which is accessible via FTP. Therefore you have to change into the “Dashboard” and search for “FTP DIAGNOSTIC LOGS”. You will be asked for the Deployment-User (which is also responsible for the Git-Deployment) as Login!

I require more Logging- / Exception information’s

Unfortunately the Error-Pages are not always constructive therefore you will quickly reach the point where you need the Logging framework.

ELMAH to the rescue!

I quickly decided for ELMAH including the additional ASP.NET MVC Integration:

Now integrate the XmlFile-Provider into the web.config:

<elmah>
  <errorLog type=”Elmah.XmlFileErrorLog, Elmah” logPath=”~/App_Data” />
</elmah>

Now you are able to navigate to the page and the App_Data directory via FTP:

Of course Log4Net or NLog would work as well.

Result:

The Azure websites act basically like a normal web application on an ordinary IIS and therefore they have almost the same rules. You don’t need a totally different repertoire like with the Cloud Services (with the Table Storage and the Diagnostics).

TL;DR:

Snapins: the old way

Snapins (like for example WebAdministration PSSnapin in the era of Windows Server 2008) are the old way which could be used in the first version of Powershell.

Disadvantages of this method:

- The PSSnapins have to be written in a .NET language and have to be available as an Assembly (there are some Sample Codes for the production of PSSnapIns in MSDN)

- The Assemblies have to be installed with a installutil.exe

What PSSnapins are installed on my system?

You will find all installed Snapins on the CmdLet “Get-PSSnapin-registered”:

Where are the PSSnapins saved? Of course in the registry!

The registration of the PSSnapins happens in the Registry:

Computer\HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\PowerShell\1\PowerShellSnapIns\

What CmdLets are responsible for the administration of the PSSnapins?

Get-PSSnapIn –registered: delivers all available PSSnapIns
Add-PSSnapIn [name]: adds the PSSnapIn to the PowerShell Session
Get-Command –pssnapin [name]: adds CmdLets to the PSSnapin (has to be added via add to the session in the beginning)

All in all you should not create Snapins anymore – but they are still in use and therefore this explanation.

PowerShell modules – the new world

The modules get established with the second version of PowerShell and they offer a lot more opportunities without the need for registration in the registry.

Where are the modules saved?

If the only task is to import a module via the module-name with Import-Module usually it takes a look into the folders which are defined in the environment variable $env:psmodulepath:

Otherwise it is also possible to define a certain pad to Powershell modules at the Import-Module.

Note: In Powershell 3.0 all available modules are activated automatically in the session and you don’t have to load them with the Import-Module.

What modules are “installed”?

You will find a list on “Get-Module –ListAvailable”:

What is the “ModuleType”?

Other then PSSnapins modules can consist of simple scripts, a workflow, an assembly or a manifest. Therefore is the definition of the ModuleType:

Binary
A module whose members are defined within an assembly (.dll extension), such as a snap-in or a class library that contains cmdlet classes. This field is introduced in Windows PowerShell 2.0.

Cim
A cmdlets-over-objects module (.cdxml extension).

Manifest
A module that is defined by a module manifest file (.psd1 extension) whose ModulesToProcesskey is empty. This field is introduced in Windows PowerShell 2.0.

Script
A module whose members are defined within a script module file (.psm1 extension). This field is introduced in Windows PowerShell 2.0.

Workflow
A workflow module (.xaml extension).

From this MSDN article.

What does such a module look like? Let’s have a look at the “WebAdministration” module

The “WebAdministration”-Module (to administrate the IIS) is located in the usual “WindowsPowershell” folder and it contains some files:

The .psd1 file is the Module Manifest and in fact what is important:

@{
GUID='{13c15630-959c-49e4-a977-758c5cc93408}'
Author='Microsoft Corporation'
CompanyName='Microsoft Corporation'
Copyright='© Microsoft Corporation. All rights reserved.'
ModuleVersion='1.0.0.0'
PowerShellVersion='3.0'
CLRVersion='4.0'
TypesToProcess='iisprovider.types.ps1xml'
FormatsToProcess='iisprovider.format.ps1xml'
NestedModules='Microsoft.IIS.PowerShell.Provider.dll'
RequiredAssemblies='Microsoft.IIS.PowerShell.Framework.dll'
AliasesToExport='Begin-WebCommitDelay','End-WebCommitDelay'
HelpInfoUri="http://go.microsoft.com/fwlink/?LinkID=216903"
}

The main functionality of the Powershell module is in the Microsoft.IIS.PowerShell.Provider.dll CLR 4.0 GAC.

How do I write a Windows Powershell Module?

After all the development of .NET based Powershell Modules is quite similar to the old Snapins:

How to Create a Powershell 2.0 Module and Cmdlet with Visual Studio 2010 (Screencast included)

A good reference is also Windows Azure CmdLets.

How do I register new modules?

The CmdLets for the registration of modules is also similar to the Snapins:

Get-Module -ListAvailable
Import-Module name
Get-Command -module name

At this place I want to give some additional information’s about the subject I’ve already written about in this article. It’s about how to show units in your code which are not given in the framework like file sizes (MB, GB), temperatures or unit of length. After I published the first article I received much feedback especially from Norbert and my coworker Erik Baum who got a little bit deeper into the subject. Erik improved the data type with some impressing features which I would like to present to you know:

Struct vs. class

The first question people are used to ask me was: Why didn’t you use a struct for the data type? I know Structs from C and for me they were like a forerunner of classes. I knew that there are also structs for C# but I didn’t know how and when I should use them.

Here the short answer:

You use structs to create types that act like integrated data types like for example DataTime. That’s exactly what we were looking for! The advantage is a lower Overhead and with that a reduced memory use. On the other hand structs are not able to do everything a class can. You can find a list in the MSDN.

Overloaded operators

If we are building a data type it would be nice if it is as easy to use as Datetime and so on. To use the operators +-/== they have to be overloaded. This is not too hard and it offers a lot of comfort for the user.

The implicit operator

The implicit operator offers the conversion of data types without the need to tell him to. With the help of this operator the following command is possible:

IComparable and IEquatable

We allow our structs to inherit from these two interfaces. If you implement them you will be able to use many cool framework features. For example the Sort function for Arrays and lists uses IComparable Interface and that’s why we are able to sort our lists like that:

The full code for the DataSize struct is available here.

.NET 4.0 offers a new Global Assembly Cache (GAC) which you can find at this path:

C:\Windows\Microsoft.NET\assembly

Every .NET 4.0 or CLR 4.0 application use this GAC.

What happens to the „old“ GAC?

The old GAC still exists on this path:

C:\Windows\assembly

It is still responsible for .NET 2.0 – .NET 3.5 or CLR 2.0 applications.

CLR? .NET Framework Version? Isn’t that the same thing?

The subject is a little more difficult but I like it as a short form:

CLR + Libraries = .NET Framework

While CLR is the base of the execution. With several .NET framework versions they added numerous libraries but there are just two big CLR Versions till today – Version 2.0 and 4.0

Source

Why this change?

At this point I want to name my source: Stackoverflow

“In .NET Framework 4.0, the GAC went through a few changes. The GAC was split into two, one for each CLR.

The CLR version used for both .NET Framework 2.0 and .NET Framework 3.5 is CLR 2.0. There was no need in the previous two framework releases to split GAC. The problem of breaking older applications in Net Framework 4.0.

To avoid issues between CLR 2.0 and CLR 4.0 , the GAC is now split into private GAC’s for each runtime.The main change is that CLR v2.0 applications now cannot see CLR v4.0 assemblies in the GAC.”

Reason for the change:

For example, if both .NET 1.1 and .NET 2.0 shared the same GAC, then a .NET 1.1 application, loading an assembly from this shared GAC, could get .NET 2.0 assemblies, thereby breaking the .NET 1.1 application

“The CLR version used for both .NET Framework 2.0 and .NET Framework 3.5 is CLR 2.0. As a result of this, there was no need in the previous two framework releases to split the GAC. The problem of breaking older (in this case, .NET 2.0) applications resurfaces in Net Framework 4.0 at which point CLR 4.0 released. Hence, to avoid interference issues between CLR 2.0 and CLR 4.0, the GAC is now split into private GACs for each runtime.”

Another point:

“Mark Miller said… June 28, 2010 12:13 PM

Thanks for the post. “Interference issues” was intentionally vague. At the time of writing, the issues were still being investigated, but it was clear there were several broken scenarios.

For instance, some applications use Assemby.LoadWithPartialName to load the highest version of an assembly. If the highest version was compiled with v4, then a v2 (3.0 or 3.5) app could not load it, and the app would crash, even if there were a version that would have worked. Originally, we partitioned the GAC under it’s original location, but that caused some problems with windows upgrade scenarios. Both of these involved code that had already shipped, so we moved our (version-partitioned GAC to another place.

This shouldn’t have any impact to most applications, and doesn’t add any maintenance burden. Both locations should only be accessed or modified using the native GAC APIs, which deal with the partitioning as expected. The places where this does surface are through APIs that expose the paths of the GAC such as GetCachePath, or examining the path of mscorlib loaded into managed code.

It’s worth noting that we modified GAC locations when we released v2 as well when we introduced architecture as part of the assembly identity. Those added GAC_MSIL, GAC_32, and GAC_64, although all still under %windir%\assembly. Unfortunately, that wasn’t an option for this release.”

This message helped me to get a clearer view on the subject. Also I didn’t know that the weird „Assembly-Viewer“ has been removed from Windows 8:

Result:

Good to know

With .NET 3.5 they’ve published two new deployment-methods for the framework:

- Full Framework

- Client Profile

Shortcut: With .NET 4.5 the construct that caused us all a lot of headache is gone.

Why all this?

The initial idea was to create a totally new and slim Framework. Other ideas might be about the security because why should a client pc provide components for server applications?

The problem

As a web- or “server” developer you are usually never get in touch with the .NET client profile. It will be a problem if you try to take some components from a “client profile” application which might have some kind of reference on the whole big framework.

You will find a nice overview about the subject on the Stackoverflow answers.

Example:

.NET 4 client profile application:

As far as I remember Visual Studio 2010 doesn’t act as clever (what lead to this errors) as Visual Studio 2012 and didn’t show this hint:

With .NET 4.5 there isn’t a client profile

“Starting with the .NET Framework 4.5, the Client Profile has been discontinued and only the full redistributable package is available. Optimizations provided by the .NET Framework 4.5, such as smaller download size and faster deployment, have eliminated the need for a separate deployment package. The single redistributable streamlines the installation process and simplifies your app’s deployment options.”

Source: MSDN

Result