MSBuild and Web Deployment Just Stops

This super helpful error message and some Googling led to a solution. I really did not believe it when I found it. But some days are like that.

Severity    Code    Description    Project    File    Line    Suppression State
Error        Web deployment task failed. ((6/29/2016 10:07:24 AM) An error occurred when the request was processed on the remote computer.) (6/29/2016 10:07:24 AM) An error occurred when the request was processed on the remote computer. Unable to perform the operation. Please contact your server administrator to check authorization and delegation settings.

And the winning blog post (certainly not MS docs) goes to A Software Guy’s Blog (Eric Kelm).

Summary
When you add Microsoft Web Deploy to an IIS server, the install adds two local user accounts, both with an expiring password. So 90 days, or whatever your expiration policy is, after you install, web deployment just stops working. The trick is to modify both accounts to check “password never expires”.

Thanks, Eric. And thanks, Microsoft. Eric’s post is four years old and this problem still exists. Epic fail.

Upgrading to .NET Core RTM and CamelCasing

This post is primarily a place holder and reminder to review Rick Strahl’s post on this topic a few more times.

The change in default serialization does not yet affect me because I was not relying on RC2 for anything serious. However it is critical to note this default behavior for future reference because many of my APIs serialize “as is” so most of my JSON is PascalCase and changing that midstream would bork every client that relies on case sensitivity in the data.

I particularly like Rick’s take on this:

“Personally I prefer to see stuff like this as an easily accessible option, and not something that is rammed down your throat - serialization shouldn't arbitrarily change the naming of properties. While that is often desirable I'm betting there are plenty of scenarios (especially when dealing with .NET to .NET) where you'd much rather have the serialized content reflect the original naming.

“It's especially frustrating in that this doesn't just affect server code where you could easily refactor, but client side code that may already be running. It also makes it harder to port existing code to ASP.NET core and it's not exactly something that's super easy to find if you're searching unless you already know that there serializer settings being sent to JSON.NET somewhere in the framework.”

Well, now the real play begins with .NET Core 1.0 RTM.

ASP.NET Web API and IDisposable Aspect with ActionFilterAttribute

I needed to apply an aspect to my ASP.NET Web API 2 services on every controller. The trick was that I needed an IDisposable object to be created before the action method executed and then disposed when the action method completed. It’s easy enough to write an ActionFilterAttribute to do something at the beginning or at the end of a controller’s action method being called. But how to you maintain state between the OnActionExecuting and the OnActionExecuted events? That’s what this little snippet does.

public class DisposableActionFilterAttribute : ActionFilterAttribute
{
  private readonly IDisposableProvider _provider;

  public DisposableActionFilterAttribute()
  {
    _provider = IoC.Instance.Container.Resolve<IDisposableProvider>();
  }

  public override void OnActionExecuting(HttpActionContext actionContext)
  {
    base.OnActionExecuting(actionContext);
    IDisposable obj = _provider.CreateDisposableObjectMethod() as IDisposable;
    actionContext.Request.Properties[Consts.DisposableObjectPropertyKey] = obj;
  }

  public override void OnActionExecuted(HttpActionExecutedContext actionExecutedContext)
  {
    base.OnActionExecuted(actionExecutedContext);
    IDisposable obj = actionExecutedContext.Request.Properties[Consts.DisposableObjectPropertyKey]
      as IDisposable;
    if (null != obj) obj.Dispose();
  }
}

And now you wire it up and presto change-o, you’ve got an IDisposable created before execution and disposed after execution. And here’s the simple wire up for global application of the attribute in the Global.asax.cs code:

public class WebApiApplication : System.Web.HttpApplication
{
  protected void Application_Start()
  {
    //..
    GlobalConfiguration.Configuration.Filters.Add(
      new DisposableActionFilterAttribute());
    //..

I do realize there are many examples of the ActionFilterAttribute and this technique, but it’s easier to search my own blog later when I need to remember how to do this. Happy Monday!

ASP.NET Core Environment Startup Conventions

I'm liking the Config[Environment](IApplicationBuilder app) method for Startup class convention

https://docs.asp.net/en/latest/fundamentals/middleware.html

And I like the Startup[Environment] class convention mentioned here

https://docs.asp.net/en/latest/fundamentals/environments.html

But I am not sure which one I like more. Some experimentation is called for.

And more research to do on writing my own middleware.

 

Castle Windsor, ASP.NET MVC and Web API in the Same Project

Many thanks to the original source by ArtisanCode on GitHub I’ve enjoyed for solving this problem. You may also enjoy Radenko’s post and Raymund’s post. I hope that you will find what I have learned from these sources helpful. The requirement was running ASP.NET MVC and Web API in the same project with a single inversion of control (IoC) container shared by both, though I never ended up using injection for my MVC controllers as I had planned.

My personal preference in the past has been Castle Windsor, in part because I like the ability to do interception so easily. More on interception in a future post. For now, have a look at how easy it is to wire up. First, we wire up the container in the Application_Start method in your Global.asax.cs class that inherits from System.Web.HttpApplication:

IoC.Instance.Container.Install(new DependencyConfig());
IoC.Instance.Container.BeginScope();	
GlobalConfiguration.Configuration.DependencyResolver = 
   new CastleDependencyResolver(IoC.Instance.Container);

Then we wire up the DependencyConfig class mentioned in the above snippet.

public class DependencyConfig : IWindsorInstaller
{
   public void Install(IWindsorContainer container, IConfigurationStore store)
   {
      //... config stuff ... blah blah blah

      //register a component with a custom interceptor for capturing some aspect
      container.Register(Component.For<IMyComponent>()
         .ImplementedBy<MyComponent>()
         .LifestyleSingleton()
         .Interceptors(InterceptorReference.ForType<MyInterceptor>())
         .Anywhere);

And implement your interceptor with something like this:

public class MyInterceptor : IInterceptor
{
   public void Intercept(IInvocation invocation)
   {
      //do something before the call is executed
      //execute the call
      invocation.Proceed();
	  //do something after the call is executed
   }
}

Now put it all together and you have ApiControllers injecting your components for you. And you can get components directly when you need them using the IoC singleton I wrote like this:

var mycomp = IoC.Instance.Container.Resolve<IMyComponent>();

Here’s the IoC singleton:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
using Castle.Windsor;

namespace MyCastle
{
   public class IoC
   {
      private static volatile IoC _instance;
      private static object _syncRoot = new object();

      private readonly WindsorContainer container;

      private IoC()
      {
         container = new WindsorContainer();
      }

      public static IoC Instance
      {
         get
         {
            if (null == _instance)
            {
               lock (_syncRoot)
               {
                  if (null == _instance)
                  {
                     _instance = new IoC();
                  }
               }
            }
            return _instance;
         }
      }

      public WindsorContainer Container { get { return container; } }
   }
}

And where does the CastleDependencyResolver and CastleDependencyScope classes come from? That’s the fun part. First, you’ll need two NuGet packages:

<?xml version="1.0" encoding="utf-8"?>
<packages>
  <package id="Castle.Core" version="3.3.3" targetFramework="net45" />
  <package id="Castle.Windsor" version="3.3.0" targetFramework="net45" />
</packages>

Then you just need the ArtisanCode code for those two classes. For my own use, I’ve modified them a bit in my own projects but you may find that you can use them right out of the box or even adopt the entire WebApiCastleIntegration library.

There you have it.

Slugs in dasBlog vs BlogEngine.NET

I realized today that my old pre-2012-04 posts were broken. That’s when I moved from dasBlog to BlogEngine.NET. At the time, I had to fix the same problem that I recently re-introduced when upgrading to a clean install of the lastest BlogEngine.NET.

  <system.webServer>
    <security>
      <requestFiltering allowDoubleEscaping="true" />
    </security>

And today I will fix my mistake by blogging about it. To support URLs with a “+” in the Title slug rather than a “-“ we need to modify the web.config as shown above.

And that makes both of these links work just fine:

http://tsjensen.com/blog/post/2016/06/08/Exploring-NET-Core-10 

http://tsjensen.com/blog/post/2010/12/03/SOLID+Principles+Of+Class+Design 

Don’t forget your configuration when you migrate or upgrade.

Bad Code Leads to Worse Code

It is 100% my fault. I wrote a small library for accessing files in Azure Storage. It has an Exists method. But rather than returning false when the path does not exist, it throws a 404 Not Found exception. This is bad code. It should just return false. And not having time to fix it, I wrote the following worse code.

try
{
   var exists = await _blobIo.Exists(fileName);
}
catch (Exception e)
{
   if (null != e.InnerException && e.InnerException.Message.Contains("404"))
   {
      fileName = $"other/{key}/{id}.{ext}";
      try
      {
         var exists = await _blobIo.Exists(fileName);
      }
      catch (Exception ie)
      {
         if (null != ie.InnerException && ie.InnerException.Message.Contains("404"))
         {
            throw new HttpResponseException(
               new HttpResponseMessage(HttpStatusCode.NotFound)
               {
                  Content = new JsonContent(new 
                  { 
                     Error = "Not Found", 
                     Message = "Content not found." 
                  })
               });
         }
      }
   }
}

I hang my head in shame. Lesson: fix the bad code. Don’t write worse code to work around the bad code.

Update
Once the fire was put out, I fixed the original library and now have this code. It's better. Could be improved, but it's not as ugly as before.

if (false == await _blobIo.Exists(fileName))
{
   fileName = $"other/{key}/{id}.{ext}";
   if (false == await _blobIo.Exists(fileName))
   {
      throw new HttpResponseException(
         new HttpResponseMessage(HttpStatusCode.NotFound)
         {
            Content = new JsonContent(new 
               { 
                  Error = "Not Found", 
                  Message = "Content not found."
               })
         });
   }
}

Second lesson: Don't let stinky code lie around too long. Clean up your messes.

Context Switching and Task<Result>.ConfigureAwait Method

While heavily involved in a few ASP.NET Web API 2 projects over the past year, I’ve learned from experience that context switching, marshaling between threads when using the async and await constructs in C# can be undesirable.

To avoid the context switching which is undeniably needed in a UI application where you’re offloading work from the UI thread, the default value of “true” for the ConfigureAwait Method on the Task<Result> class, one must set the “continueOnCapturedContext” to false. Like this:

var user = await repo.GetUser(id).ConfigureAwait(false);

There is some debate as to whether this is helpful in an ASP.NET application given the request is handled on a thread pool thread which is returned to the pool while the asynchronous task is running.

The issue I’ve noticed is that in some cases setting the “continueOnCapturedContext” to false, eliminates the need to marshal the continuation of your code back to the original context and use of another thread pool thread. While there are no noticeable performance advantages, I have noticed that I experience fewer threading exceptions when I allow the execution to continue on the thread that just executed my asynchronous work.