The Way I See It

BizTalk–identifier ‘EDI’ does not exist in ‘Orchestration’; are you missing an assembly reference?

Shinva — Wed, 08 May 2013 09:10:38 GMT

While working with EDI messages in Orchestrations you might get the following error message.

Identifier ‘EDI’ doesn't exist in ‘Orchestration’; are you missing an assembly reference?

This happens when you try to access the EDI properties of a message in an orchestration like EDI.ISA06

To resolve this issue ensure the Orchestration Project has a reference to the assembly Microsoft.BizTalk.Edi.BaseArtifacts.dll

It is located in the folder you installed BizTalk Server ( C:\Program Files (x86)\Microsoft BizTalk Server 2010\ )

AppFabric Cache–How to check if a region exists

Shinva — Tue, 30 Apr 2013 04:52:12 GMT

In an AppFabric Cache you can segregate your cached data into regions. You create a region using the CreateRegion() method of the DataCache object.

Working on a recent project I was asked How do I check if a cache exits ? There is no RegionExits() method on the DataCache object.

To check if a region exists you can make use of the CreateRegion() method.

The CreateRegion returns a bool. As per the documentation a value of true indicates that the region created successfully. A value of false indicates that the region already exists.

REF : http://msdn.microsoft.com/en-us/library/microsoft.applicationserver.caching.datacache.createregion(v=azure.10).aspx

Tools To Simulate CPU / Memory / Disk Load

Shinva — Sat, 27 Oct 2012 05:19:12 GMT

I have seen a lot of people searching for tools to simulate high CPU, Memory and Disk issues. Here are the tools I use

Simulate High CPU Usage

CPUSTRES.EXE is a tool you can use to simulate High CPU usage by an user mode process. Its available for download from the Windows Sysinternals website.

Here I have used CPUSTRES to simulate 50% CPU usage.

CPUSTRES also has options to change the priority of the threads it spawns.

Simulate High Memory Usage

To simulate high memory usage you can use the TestLimit tool from the Sysinternals website. TestLimit can be used to simulate a variety of memory leak issues. Here I am using TestLimit to reduce the Available memory on my machine.

Simulate High Disk Activity

SQLIO is a tool for benchmarking the I/O capacity of a given storage system. It can simulate sequential and random I/O and is usually used to test storage systems for SQL Server installations.

SQLIO is an useful tool that you can use to simulate high disk activity.

Simulate High Disk Usage

Another useful tool to simulate these issues is consume.exe. It ships with Windows SDK and can be used to consume resources like CPU, memory and disk. Here I am using consume.exe to simulate low disk space issues.

NOTE : Use these tools with extreme caution as they can freeze the machine you run them on and you might end up rebooting the machine.

Where is ASP.NET 4.5 …wait Where is .NET 4.5 ?

Shinva — Fri, 12 Oct 2012 10:17:44 GMT

When .NET 3.5 was released a lot of people wondered why ASP.NET 3.5 wouldn’t show up in IIS. http://blogs.msdn.com/b/vijaysk/archive/2008/03/20/running-asp-net-3-5-on-iis.aspx

Well with .NET 4.5 you might be a bit more confused…

So first .NET 4.5 will not show up in IIS

But now if you check the Microsoft.NET framework folder you will see that you will not have a .NET 4.5 folder as well

Well Why ?

If you have read my previous post by now you would have understood that there are two ways the .NET framework is upgraded.

1. Side By Side release – Like v1.1 and v2.0. There releases are completely independent of each other

2. Enhancements – Like v3.0 and 3.5

.NET 3.5 and .NET 3.0 are just additions to the .NET 2.0. So the v2.0 folder has all the .NET 2.0 files and v3.0 and v3.5 folders have all the files that are required for the enhancements like WCF, LINQ. If you wanted to use these enhancements in ASP.NET your web.config files had to explicitly have references to these 3.5 assemblies.

.NET 4 is a Side By Side upgrade. Which means it can exist independent of v1.1 and v2.0.

.NET 4.5 is an enhancement, but unlike 3.0 or 3.5 it will not be separated out. It is an in place upgrade.

Which means once you install .NET 4.5 the v4.0 folder will be updated to contain all the .NET 4.5 files.

Does that mean you wiped out .NET 4.0 from your machine ?

Well yes and no.

Yes, because the installation updates the v4.0 folder to .NET 4.5

My machine has both VS 2010 and VS 2012 installed. If I launch the VS 2010 Command Prompt and run launch the C Sharp compiler it will say 4.5. There is no compiler for 4.0 after the upgrade.

No, because even though you now have a single updated folder, you can control which version of .NET 4 your application will use.

Visual Studio 2012 provides you an option to target either .NET 4 or .NET 4.5

When you switch the Target framework in Visual Studio two thing happen

1. The config file will reflect the targetFramework

A .NET 4.5 web.config will have an entry similar to

and if you choose .NET 4 it will be

Similarly in a Windows Application the version is controlled in its app.config file with

2. The reference assemblies for the common namespaces like System are updated.

The reference assemblies are still separate for .NET 4 and .NET 4.5

Do you have to upgrade your existing ASP.NET 4 web application to .NET 4.5 because of this?

No, if you have an ASP.NET web application built using VS 2010, it will have a compilation tag with targetFramework="4.0" in its web.config already.

Which means it will continue to work fine even after the .NET 4.5 upgrade.

In case you want to use the new .NET 4.5 features like async that's when you will have to upgrade your web application.

IIS 8 What's new – Website settings

Shinva — Thu, 11 Oct 2012 07:06:48 GMT

The two new additions in IIS 8 are

Preload Enabled

This setting is available if you have the Application Initialization module installed

The preloadEnabled metabase setting along with the startMode setting can be used to ‘warm up’ your web application.

When you set the startMode property of your application pool to AlwaysRunning a worker process is spawned as soon as IIS starts up and does not wait for the first user request. But this does not mean the web application is initialized.

When you set preloadEnabled to true, IIS will simulate a user request to the default page (can be changed with initializationPage metabase setting) of the website/virdir so that the application initializes. The request is not logged in the IIS logs.

But you can trace this with FREB. Every time you restart your application pool you will see a FREB trace file for the dummy request. You can identify this request by analysing the GENERAL_REQUEST_HEADERS and looking at the User-Agent string.

User-Agent: IIS Application Initialization Preload

Maximum Url Segments

With the maxUrlSegments metabase setting you can control the number of segments in an URL that your web application can serve. A segment is nothing but the number of / in your URL. This is a security setting that you can use to control the depth to which a user can browse your website.

So for example say you have an application whose URL are mostly of the format http://website/virdir/page, you can set the maxUrlSegments to 3. This stops probing attacks.

If a user tries to browse an URL with segments exceeding this limit he/she will see a 404 message and a 404.20 HTTP status code will be logged in the IIS logs.

IIS 8 What's new – Application pool settings

Shinva — Wed, 10 Oct 2012 06:15:05 GMT

Here is a list of enhancements to Application Pools in IIS 8.

Start Mode

A few web applications take a significant amount of time to start up. IIS by default only launches a worker process when the first request for the web application is received. So for the web applications that require a longer time to initialize, users might see slow responses.

For such applications it is a good idea to launch the worker process as soon as IIS is started. The application pools have a startMode setting which when set to AlwaysRunning launches the worker process for the application pool as soon as IIS is started.

IIS 8 provides you this setting in the Application Pool Settings UI.

This feature is also available in IIS 7.5, but you will have to edit the applicationHost.config directly to enable this feature.

CPU Throttle

IIS 7.5 and earlier have a CPU monitoring feature that you could use to terminate a worker process, when its CPU usage goes beyond a certain limit. Its very useful in shared hosting scenarios.

In IIS 8 this feature has been enhanced. You can now instead of terminating the worker process, throttle it.

The KillW3wp will terminate the worker process like IIS 7.5. But if set to Throttle or ThrottleUnderLoad instead of terminating the worker process it will limits its CPU usage to the value set.

This is an application pool setting, so the limit you set applies to all the worker processes (web garden) of an application pool. So for an application pool with 5 worker processes, if the value for limit is set to 50000 the combined CPU usage of all the 5 worker processes is throttled at 50%.

Process Affinity Mask (64 Bit)

The UI now also exposes the smpProcessorAffinityMask2 metabase key. This is used when you enable the Processor Affinity setting.

Generate Process Model Event Log Entries

The logEventOnProcessModel metabase setting controls if an event has to be written when an application pool is shutdown because it was idle.

When set to True an event similar to the one below is logged in the System Event Log when WAS shutdown the application pool.

Windows 8 is the Super workstation OS I had been waiting for

Shinva — Sat, 06 Oct 2012 11:45:50 GMT

Long back I had written a blog post about Using Windows Server 2008 as a SUPER workstation OS that caused a lot of buzz

Server 2008: The Windows Workstation we always wanted

Review: Using Windows Server 2008 on a PC

I had my reasons to use a Server OS on my work laptop and have been doing it since Windows Server 2003. When I wrote that blog post I had a vision of Windows which I now think has become a reality.

After nearly 9 years I have made my switch to a desktop version of Windows. My work laptop now has Windows 8 (RTM).

And before you drift into the world of Touch and Apps. My laptop doesn't have a Touch Screen. I am still a mouse and keyboard guy.

Search and Launch is way better than the Start Button

I am surprised at the number of hard core Start Button fans. The most common use of the Start Button was to launch apps. I had ditched the Start Menu long back for the Search Bar. Simply because my “All Programs” list had grown to three columns.

To launch any program I would just hit Start and type the first few characters of the program I wanted to run. Say you want to launch Paint … Just hit the Windows button and type in “pa..” and hit Enter.

With Windows 8 it becomes better… You can search Apps, Settings and Files in one place. You can also launch search within apps. Like search within the Store App.

I have seen a lot of people use Search to launch Apps than actually traversing the “All Programs” list.

And in case you can’t do without it here is a tip I learnt. You can create a Custom Toolbar and point it to C:\ProgramData\Microsoft\Windows\Start Menu\Programs. You will get a list similar to “All Programs”

Searching and Launching Apps is way faster than clicking the Start Button and traversing the list. And its not just for power users. Say you want to change the Sounds your computer makes. Now instead of knowing that you have to open the Control Panel you can just hit the Windows button and type in Sound. You will be given a list of Apps and Settings related to Sound. Go to the Settings section and you can now easily change the Sound Settings.

The most unexpected outcome of the new Start Screen for me … Its strange but a Start Screen that covers the entire screen is something I didn't realize I was missing all these days. There have been a lot of instances when I have been working on something confidential and a colleague walks by your desk. I am pretty sure a lot of people have been in this situation. You struggle to hit the Minimize button with your mouse or try to switch to some other app or hit Ctrl+M or (and I have seen this happen) just turn off the screen. Now I just hit the Windows Key and voila the entire screen is covered.

And the concept of Tiles which display updates is also very useful.

Virtualization ! … Virtualization on my workstation is a must have for me. A lot of people think virtualization is not something meant for a client OS.

Fist it enables me to break and restore as many times as I need. I now have a Windows 2008 R2 server running virtualized on my Windows 8 machine. I can do all my risky experimentation on my VPC and even if I end up rendering the VPC unusable I can just restore back to an earlier snapshot.

Since the host machine can be networked to the virtual PCs you can partition your software. Move all your heavy software like SQL Server to a virtual PC and it can be access by the host machine like a regular networked machine. The advantage ?… you can save /turn off the virtual PC releasing all the resources when not in use.

You no longer have to deal with the limitations of Virtual PC or Virtual Box you get Hyper V with Windows 8.

One problem with enabling Hyper V on Windows 2008 R2 was that the machine would no longer Sleep or Hibernate. With Windows 8 there are no such restrictions. I can just close my laptop lid and even with Virtual Machines running, the laptop goes into Sleep mode. You lift the lid and your virtual machines are up and running where you left them.

For Hyper V you would need the Pro / Enterprise edition and a machine with SLAT.

Apps are not just for phones and tablets… Initially you might be averse to Apps, the concept of having two different types of applications might sound a bit strange. For me the concept of Apps is simple … its controlled software. The last time you installed a game from the internet did you bother to check if it used your location information or connected to the network ? Not really right … now think of all the consequences because of that. When you download an app from the Store it clearly warns you about what the app can access … like use your internet connection. The Apps run in a Sandbox and that way are very safe. We all download small apps like games or utilities from unknown sources… doing it from the Store is safer as all Apps in the Store are verified.

For me the most used App is the Reader. You no longer have to install a software for reading PDFs. I no longer have to choose between Adobe and FoxIt … I love the simplicity of the Reader.

Multitask Without Interruptions…The Snap View is very useful for multitasking. You can keep an eye on the stock market while doing you regular work on your Desktop.

The Duplicates… One thing that you might find strange is that you now have two Internet Explorers. But what I have seen is that depending on who uses the computer eventually one will take over. For me the Desktop IE is the preferred choice because some of our corp websites require ActiveX plugins. But on my home PC the new IE gets used a lot. The new IE is better both in terms of presentation and safety. Having duplicates is not a big deal … a lot of people install multiple browsers and media players on their desktop but eventually.

The simple things like Using Corners instead of button clicks saves you a lot of time. And its doesn't take long to adapt to it.

One feature I always missed was the ability to mount ISOs. Now with Windows 8 you can mount ISO files and VHDs easily.

Windows 8 has a lot of features in it for both the power user and a normal user.

Its Smarter than you think… I was pleasantly surprised by what Windows 8 does in the background. I have been using Windows 8 for nearly two months now and my disks have 0 fragmentation… which basically means better performance. The best part is I never knew when it defragmented those disks (until of course I looked at the Last Run column) because it never interrupted my work and carried out this maintenance work when the machine was idle.

Even without all the Touch goodness Windows 8 still puts on a great show.

BizTalk–Map Incoming Message to a string Field

Shinva — Tue, 02 Oct 2012 16:40:18 GMT

In a recent project I worked, one of the requirements was to copy the entire incoming message to a String Field.

Consider the incoming message

<ns0:TestNode xmlns:ns0="http://testEnvelope.SourceSchema">
< FieldOne>FieldOne_0</FieldOne>
< FieldTwo>FieldTwo_0</FieldTwo>
< /ns0:TestNode>

This would have to be mapped to

<ns0:Root xmlns:ns0="http://testEnvelope.DestinationSchema">
< IncomingMsg>IncomingMessageXML</IncomingMsg>
< /ns0:Root>

so the output message should look like

<ns0:Root xmlns:ns0="http://testEnvelope.DestinationSchema">
< IncomingMsg><ns0:TestNode xmlns:ns0="http://testEnvelope.SourceSchema"> <FieldOne>FieldOne_0</FieldOne> <FieldTwo>FieldTwo_0</FieldTwo> </ns0:TestNode></IncomingMsg>
< /ns0:Root>

In the BizTalk Mapper you cannot just map a node to the output field.

Well you can but the output is always going to be an empty field as shown below.

<ns0:Root xmlns:ns0="http://testEnvelope.DestinationSchema">
<IncomingMsg/>
< /ns0:Root>

To achieve this you will have to make use of Scripting Functoids as follows

The first Functoid defines a C# function that takes a NodeIterator as the input and returns the OuterXml.

The second Functoid uses XSLT to map the incoming message to the string field called IncomingMsg in the output message. To achieve this it calls the NodeToString function that is defined in the previous Functoid. We pass it /TestNode which is the root node of the incoming message.

Since the first Functoid is just being used as a place holder to hold the C# function you will receive the following warning when you test the map in Visual Studio. But it can be ignored.

RFC_ERROR_LOGON_FAILURE. SapErrorMessage=Name or password is incorrect (repeat logon)

Shinva — Thu, 27 Sep 2012 11:45:31 GMT

While trying to connect to a SAP server from Visual Studio - Consume Adapter Service, you might encounter the following error

Microsoft.ServiceModel.Channels.Common.ConnectionException: Details: ErrorCode=RFC_OK. ErrorGroup=RFC_ERROR_LOGON_FAILURE. SapErrorMessage=Name or password is incorrect (repeat logon). AdapterErrorMessage=. ---> Microsoft.Adapters.SAP.RFCException: Details: ErrorCode=RFC_OK. ErrorGroup=RFC_ERROR_LOGON_FAILURE. SapErrorMessage=Name or password is incorrect (repeat logon). AdapterErrorMessage=.

This need not be a username/password issue. The BizTalk SAP Adapter uses the SAP RFC SDK to communicate with the SAP Server. There are different version available 6.4, 7.0 and 7.1

Looking at the versions, the machine on which we had this issue had the 6.4 version of the SDK. You can figure out the version of the SAP RFC SDK by looking at the version of librfc32u.dll

We upgraded the SDK to version 7.1 and were able to connect to the SAP Server.

.NET 4.5 - Information of Caller Function (Caller Attributes in .NET 4.5)

Shinva — Thu, 27 Sep 2012 06:40:03 GMT

While debugging code “Who called my function ?” is a million dollar question. Knowing the origin of your function call is in many cases the first step in debugging any code. Until now a few ways of doing this were to look at the CallStack in visual studio or a debugger or the most common Exception.StackTrace. .NET 4.5 has added a new feature with which a function can now know who is calling it. The function being invoked can now know the origin of its call … that would include the Function name, File name and the Line number from where the call originated. Developers can now tap this information to provide useful debugging information. These features are part of the System.Runtime.CompilerServices namespace. Here is a sample and its output

So in MyTestFunction you can now see that the Main function in Program.cs called it from Line 14. The developer can log this as additional information for debugging.

You do this by defining three parameters for your function that will hold your Caller data. You then use the CallerMemberName, CallerFilePath and CallerLineNumber attributes which are part of the System.Runtime.CompilerServices namespace to decorate the arguments of your function.

And in your function these three parameters will now have the Caller function details populated each time the function is invoked.

The way this is implemented is using the Optional parameters feature which was introduced in .NET 4. Take a look at the MSIL for the above code

The three parameters that we decorated using the Caller attributes are marked as Optional and in the beginning of the function are loaded with the Caller details.

Which basically means you can also do this.

Here instead of leaving those optional parameters to the defaults I am explicitly passing them values. And the output will display the values I passed.

Just to show that these are normal optional parameters. And be aware that the developer of the caller function can accidently or intentionally overwrite these values.

This feature need not be limited to just providing debugging information you can also take decisions based on the Caller

A very useful feature in .NET 4.5 where your functions can now know details of the Caller function.