Scott Hanselman

13 hours debugging a segmentation fault in .NET Core on Raspberry Pi and the solution was...

July 18, '17 Comments [60] Posted in Bugs | DotNetCore
Sponsored By

Debugging is a satisfying and special kind of hell. You really have to live it to understand it. When you're deep into it you never know when it'll be done. When you do finally escape it's almost always a DOH! moment.

I spent an entire day debugging an issue and the solution ended up being a checkbox.

NOTE: If you get a third of the way through this blog post and already figured it out, well, poop on you. Where were you after lunch WHEN I NEEDED YOU?

I wanted to use a Raspberry Pi in a tech talk I'm doing tomorrow at a conference. I was going to show .NET Core 2.0 and ASP.NET running on a Raspberry Pi so I figured I'd start with Hello World. How hard could it be?

You'll write and build a .NET app on Windows or Mac, then publish it to the Raspberry Pi. I'm using a preview build of the .NET Core 2.0 command line and SDK (CLI) I got from here.

C:\raspberrypi> dotnet new console
C:\raspberrypi> dotnet run
Hello World!
C:\raspberrypi> dotnet publish -r linux-arm
Microsoft Build Engine version for .NET Core

raspberrypi1 -> C:\raspberrypi\bin\Debug\netcoreapp2.0\linux-arm\raspberrypi.dll
raspberrypi1 -> C:\raspberrypi\bin\Debug\netcoreapp2.0\linux-arm\publish\

Notice the simplified publish. You'll get a folder for linux-arm in this example, but could also publish osx-x64, etc. You'll want to take the files from the publish folder (not the folder above it) and move them to the Raspberry Pi. This is a self-contained application that targets ARM on Linux so after the prerequisites that's all you need.

I grabbed a mini-SD card, headed over to and downloaded the latest Raspbian image. I used - a lovely image burner for Windows, Mac, or Linux - and wrote the image to the SD Card. I booted up and got ready to install some prereqs. I'm only 15 min in at this point. Setting up a Raspberry Pi 2 or Raspberry Pi 3 is VERY smooth these days.

Here's the prereqs for .NET Core 2 on Ubuntu or Debian/Raspbian. Install them from the terminal, natch.

sudo apt-get install libc6 libcurl3 libgcc1 libgssapi-krb5-2 libicu-dev liblttng-ust0 libssl-dev libstdc++6 libunwind8 libuuid1 zlib1g

I also added an FTP server and ran vncserver, so I'd have a few ways to talk to the Raspberry Pi. Yes, I could also SSH in but I have a spare monitor, and with that monitor plus VNC I didn't see a need.

sudo apt-get pure-ftpd

Then I fire up Filezilla - my preferred FTP client - and FTP the publish output folder from my dotnet publish above. I put the files in a folder off my ~\Desktop.

FTPing files

Then from a terminal I

pi@raspberrypi:~/Desktop/helloworld $ chmod +x raspberrypi

(or whatever the name of your published "exe" is. It'll be the name of your source folder/project with no extension. As this is a self-contained published app, again, all the .NET Core runtime stuff is in the same folder with the app.

pi@raspberrypi:~/Desktop/helloworld $ ./raspberrypi 
Segmentation fault

The crash was instant...not a pause and a crash, but it showed up as soon as I pressed enter. Shoot.

I ran "strace ./raspberrypi" and got this output. I figured maybe I missed one of the prerequisite libraries, and I just needed to see which one and apt-get it. I can see the error, but that's a historical Debian-ism and more of a warning than a fatal.

strace on a bad exe in Linux

I used to be able to read straces 20 years ago but much like my Spanish, my skills are only good at Chipotle. I can see it just getting started loading libraries, seeking around in them, checking file status,  mapping files to memory, setting memory protection, then it all falls apart. Perhaps we tried to do something inappropriate with some memory that just got protected? We are dereferencing a null pointer.

Maybe you can read this and you already know what is going to happen! I did not.

I run it under gdb:

pi@raspberrypi:~/Desktop/WTFISTHISCRAP $ gdb ./raspberrypi 
GNU gdb (Raspbian 7.7.1+dfsg-5+rpi1) 7.7.1
Copyright (C) 2014 Free Software Foundation, Inc.
This GDB was configured as "arm-linux-gnueabihf".
"/home/pi/Desktop/helloworldWRONG/./raspberrypi1": not in executable format: File truncated

Ok, sick files?

I called Peter Marcu from the .NET team and we chatted about how he got it working and compared notes.

I was using a Raspberry Pi 2, he a Pi 3. Ok, I'll try a 3. 30 minutes later, new SD card, new burn, new boot, pre-reqs, build, FTP, run, SAME RESULT - segfault.


Maybe corruption? Here's a thread about Corrupted Files on Raspbian Jesse 2017-07-05! That's the version I have. OK, I'll try the build of Raspbian from a week before.

30 minutes later, burn another SD card, new boot, pre-reqs, build, FTP, run, SAME RESULT - segfault.



Maybe a bad nuget.config? No.

Bad daily .NET build? No.


Ok, I'll try Ubuntu Mate for Raspberry Pi. TOTALLY different OS.

30 minutes later, burn another SD card, new boot, pre-reqs, build, FTP, run, SAME RESULT - segfault.

What's the common thread here? Ok, I'll try from another Windows machine.

SAME RESULT - segfault.

I call Peter back and we figure it's gotta be prereqs...but the strace doesn't show we're even trying to load any interesting libraries. We fail FAST.

Ok, let's get serious.

We both have Raspberry Pi 3s. Check.

What kind of SD card does he have? Sandisk? Ok,  I'll use Sandisk. But disk corruption makes no sense at that level...because the OS booted!

What did he burn with? He used Win32diskimager and I used Etcher. Fine, I'll bite.

30 minutes later, burn another SD card, new boot, pre-reqs, build, FTP, run, SAME RESULT - segfault.

He sends me HIS build of a HelloWorld and I FTP it over to the Pi. SAME RESULT - segfault.

Peter is freaking out. I'm deeply unhappy and considering quitting my job. My kids are going to sleep because it's late.

I ask him what he's FTPing with, and he says WinSCP. I use FileZilla, ok, I'll try WinSCP.

WinSCP's New Session dialog starts here:

SFTP is Default

I say, WAIT. Are you using SFTP or FTP? Peter says he's using SFTP so I turn on SSH on the Raspberry Pi and SFTP into it with WinSCP and copy over my Hello World.


Hello World on a Raspberry Pi


I make a folder called Good and a folder called BAD. I copy with FileZilla to BAD and with WinSCP to GOOD. Then I run a compare. Maybe some part of .NET Core got corrupted? Maybe a supporting native library?

pi@raspberrypi:~/Desktop $ diff --brief -r helloworld/ helloworldWRONG/
Files helloworld/raspberrypi1 and helloworldWRONG/raspberrypi1 differ

Wait, WHAT? The executable are different? One is 67,684 bytes and the bad one is 69,632 bytes.

Time for a  visual compare.

All the ODs are gone

At this point I saw it IMMEDIATELY.

0D is CR (13) and 0A is LF (10). I know this because I'm old and I've written printer drivers for printers that had both carriages and lines to feed. Why do YOU know this? Likely because you've transferred files between Unix and Windows once or thrice, perhaps with FTP or Git.

All the CRs are gone. From my binary file.


I went straight to settings in FileZilla:

Treat files without extensions as ASCII files

See it?

Treat files without extensions as ASCII files

That's the default in FileZilla. To change files that are just chilling, minding their own business, as ASCII, and then just randomly strip out carriage returns. What could go wrong? And it doesn't even look for CR LF pairs! No, it just looks for CRs and strips them. Classy.

In retrospect I should have used known this, but it wasn't even the switch to SFTP, it was the switch to an FTP program with different defaults.

This bug/issue whatever burned my whole Monday. But, it'll never burn another Monday, Dear Reader, because I've seen it before now.


Why does experience matter? It means I've failed a lot in the past and it's super useful if I remember those bugs because then next time this happens it'll only burn a few minutes rather than a day.

Go forth and fail a lot, my loves.

Oh, and FTP sucks.

Sponsor: Thanks to Redgate! A third of teams don’t version control their database. Connect your database to your version control system with SQL Source Control and find out who made changes, what they did, and why. Learn more

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

facebook twitter subscribe
About   Newsletter
Sponsored By
Hosting By
Dedicated Windows Server Hosting by SherWeb

Porting a 15 year old .NET 1.1 Virtual CPU Tiny Operating System school project to .NET Core 2.0

July 2, '17 Comments [10] Posted in DotNetCore | Learning .NET
Sponsored By

The 2002 TinyOS in C# is now on .NET Core in 2017 running on UbuntuI've had a number of great guests on the podcast lately. One topic that has come up a number of times is the "toy project." I've usually kept mine private - never putting them on GitHub - Somewhat concerned that people would judge me and my code. However, hypocrite that am (aren't we all?) I have advocated that others put their "Garage Sale Code" online. So here's some crappy code. ;)

The Preamble

While I've been working as an engineer for 25 years this year, I didn't graduate from school with a 4 year degree until 2003 - I just needed to get it done, for myself. I was poking around recently and found my project from OIT's CST352 "Operating Systems" class. One of the projects was to create a "Virtual CPU and OS." This is kind of a thought exercise. It's not really a parser/lexer - although there is both - and it's not a real OS. But it needs to be able to take in a made-up quasi-Assembly Language instruction set and execute them on a virtual CPU while managing virtual memory of arbitrary size. Again, a thought exercise made real to confirm that the student understands the responsibilities of a CPU.

Here's an example "application." Confused yet? Here's the original spec I was given in 2002 that includes the 36 instructions the "CPU" should understand. It has 10 general-purpose 32bit registers address as 1 through 10. Register 10 is the stack pointer. There are two bit flag registers - sign flag and zero flag.

Instructions are "opcode arg1 arg2" with constants prefixed with "$."

11 r8        ;Print r8
6 r1 $10 ;Move 10 into r1
6 r2 $6 ;Move 6 into r2
6 r3 $25 ;Move 25 into r3
23 r1 ;Acquire lock in r1 (currently 10)
11 r3 ;Print r3 (currently 25)
24 r1 ;Release r1 (currently 10)
25 r3 ;Sleep r3 (currently 25)
11 r3 ;Print r3 (currently 25)
27 ;Exit

I write my homework assignment in 2002 in the idiomatic C# of the time on .NET 1.1. That means no Generics<T> - I had to make my own strongly typed collections. That means C# has dozens of (if not a hundred) language and syntax improvements. I didn't use a Unit Testing Framework as TDD was just starting around 1999 during the XP (eXtreme Programming) days and NUnit was just getting start. It also uses "unsafe" to pin down memory in a few places. I'm sure there are WAY WAY WAY better and more sophisticated ways to do this today in idiomatic C# of 2017. Those are excuses, the real reasons are my own ignorance, ability, combined with some night-school laziness.

One of the more fun parts of this exercise was moving from physical memory (a byte array as I recall) to a full-on Memory Manager where each Process thought it could address a whole bunch of Virtual Memory while actual Physical Memory was arbitrarily sized. Then - as a joke - I would swap out memory pages as XML! ;) Yes, to be clear, it was a joke and I still love it.

You can run an "app" by passing in the total physical memory along with the text file containing the program, but you can also run an arbitrary number of programs by passing in an arbitrary number  of text files! The "TinyOS" will handle each process thinking it has its own memory and will time

If you are more of a visual learner, perhaps you'd prefer this 20-slide PowerPoint on this Tiny CPU that I presented in Malaysia later that year. You dig those early 2000-era slides? I KNOW YOU DO.

Tiny OS Memory SlidesTiny OS Memory SlidesTiny OS Memory Slides 

Updating a .NET 1.1 app to cross-platform .NET Core 2.0

Step 1 was to download the original code from my own blog. ;) This is also Reason #4134 why you should have a blog.

I decided to use Visual Studio 2017 to upgrade it, and even worse I decided to use .NET Core 2.0 which is currently in Preview. I wanted to use .NET Core 2.0 not just because it's cross-platform but also because it promises to have a pretty large API surface area and I want this to "just work." The part about getting my old application running on Linux is going to be awesome, though.

Visual Studio then pops a scary dialog about upgrading files. NOTE that another totally valid way to do this (that I will end up doing later in this blog post) is to just make a new project and move the source files into it. Natch.


Visual Studio says it's targeting .NET 2.0 Full Framework, but I ratchet it up to 4.6 to see what happens. It builds but with a bunch of errors about Obsolete methods, the most interesting one being this one:

Warning CS0618    
'ConfigurationSettings.AppSettings' is obsolete:
'This method is obsolete, it has been replaced by
C:\Users\scott\Downloads\TinyOSOLDOLD\OS Project\CPU.cs 72

That's telling me that my .NET 1/2 API will work but has been replaced in .NET 4.x, but I'm more interested in .NET Core 2.0. I could make my EXE a LIB and target .NET Standard 2.0 or I could make a .NET Core 2.0 app and perhaps get a few more APIs. I didn't do a formal analysis with the .NET Portability Analyzer but I will add that to the list of Things To Do. I may be able to make a library that works on an iPhone - a product that didn't exist when I started this assignment. That would be Just Cool(tm).

I decided to just make a new empty .NET Core 2.0 app and copy the source .cs files into it. A few interesting things.

  • My app also used "unsafe" code (it pins memory down and accesses it directly).
  • It has extensive inline documentation in comments that I used to use NDoc to make a CHM Help file. I'd like that doc to turn into HTML at some point.
  • It also has an appsettings.json file that needs to get copied to the output folder when it compiles.
  • While I could publish it to a self-contained .NET Core exe, for now I'm running it like this in my test batch files - example:
    • dotnet netcoreapp2.0/TinyOSCore.dll 512 scott13.txt

Here's the resulting csproj file.

<Project Sdk="Microsoft.NET.Sdk">



<None Remove="appsettings.json" />

<Content Include="appsettings.json">

<PackageReference Include="Microsoft.Extensions.Configuration" Version="2.0.0-preview2-final" />
<PackageReference Include="Microsoft.Extensions.Configuration.Json" Version="2.0.0-preview2-final" />
<PackageReference Include="Microsoft.Extensions.DependencyInjection" Version="2.0.0-preview2-final" />
<PackageReference Include="Microsoft.Extensions.Options.ConfigurationExtensions" Version="2.0.0-preview2-final" />


Other than the obsolete configuration warning and a few malformed XML comments, the app compiled and ran! You can actually "watch" the nightmare process here in the form of GitHub commits. I also moved the docs from a 2002 Word Doc to Markdown so be sure to explore the fairly extensive spec

The only significant change was loading the config. Configuration is even more different on .NET Core 2.0 than Full Framework. It's FAR more, ahem, configurable. I could have used "Options," I could have written my own config provider if it was important to keep the file format.

This little TinyOS has a bunch of config options that come in from a .exe.config file in XML like this (truncated):

Must be a factor of 4
This is the total Physical Memory in bytes that the CPU can address.
This should not be confused with the amount of total or addressable memory
that is passed in on the command line.
<add key="PhysicalMemory" value="128" />
Must be a factor of 4
This is the ammount of memory in bytes each process is allocated
Therefore, if this is 256 and you want to load 4 processes into the OS,
you'll need to pass a number > 1024 as the total ammount of addressable memory
on the command line.
<add key="ProcessMemory" value="384" />
<add key="DumpPhysicalMemory" value="true" />
<add key="DumpInstruction" value="true" />
<add key="DumpRegisters" value="true" />
<add key="DumpProgram" value="true" />
<add key="DumpContextSwitch" value="true" />
<add key="PauseOnExit" value="false" />

I have a few choices. I could make a Configuration Provider and reach .NET Core to read this format (there's an XML adapter, in fact) or make the code porting easier by moving these "name/value" pairs to a JSON file like this:

"PhysicalMemory": "128",
"ProcessMemory": "384",
"DumpPhysicalMemory": "true",
"DumpInstruction": "true",
"DumpRegisters": "true",
"DumpProgram": "true",
"DumpContextSwitch": "true",
"PauseOnExit": "false",
"SharedMemoryRegionSize": "16",
"NumOfSharedMemoryRegions": "4",
"MemoryPageSize": "16",
"StackSize": "16",
"DataSize": "16"

This was just a few minutes of search and replace to change the XML to JSON. I could have also written a little app or shell script. By changing the config (rather than writing an adapter) I could then keep the code 99% the same.

My code was doing things like this (all over...there was no DI container yet):

bytesOfPhysicalMemory = uint.Parse(ConfigurationSettings.AppSettings["PhysicalMemory"]);

And I'd like to avoid major refactoring - yet. I added this bit of .NET Core configuration at the top of the EntryPoint and saved away an IConfigurationHost:

var builder = new ConfigurationBuilder()
Configuration = builder.Build();

I've got a Dictionary in the format of the IConfiguration host called "Configuration." So now I just do this in a dozen places and the app compiles again:

bytesOfPhysicalMemory = uint.Parse(Configuration["PhysicalMemory"]);

This brings up that feeling we all have when we look at old code - especially our own old code. I should have abstracted that away! Why didn't I use an interface? Why so many statics? What was I thinking?

We can beat ourselves up or we can feel good about ourselves and remember this. The app worked. It still works. There is value in it. I learned a lot. I'm a better programmer now. I don't know how far I'll take this old code but I had a lovely afternoon porting it to .NET Core 2.0 and I may refactor the heck out if it or I may not.

TinyOS on Ubuntu

For now I did update the smoke tests to run on both Windows and Linux and I'm happy with the experiment.

Related Links

Have YOU done a project like this, either in school or on your own?

Sponsor: Check out JetBrains Rider: a new cross-platform .NET IDE. Edit, refactor, test, build and debug ASP.NET, .NET Framework, .NET Core, or Unity applications. Learn more and get access to early builds!

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

facebook twitter subscribe
About   Newsletter
Sponsored By
Hosting By
Dedicated Windows Server Hosting by SherWeb

Speed of dotnet run vs the speed of dotnet for published apps (plus self-contained .NET Core apps)

June 28, '17 Comments [8] Posted in DotNetCore
Sponsored By

The .NET Core team really prides themselves on performance. However, it's not immediately obvious (as with all systems) if you just do Hello World as a developer. Just today I was doing a Ruby on Rails app in Development Mode with mruby - but that's not what you'd go to production with.

Let's look at a great question I got today on Twitter.

Dotnet Run - Builds and Runs Source Code in Development

That's a great question. If you install .NET Core 2.0 Preview - this person is on a Mac, but you can use Linux or Windows as well - then do just this:

$ dotnet new console
$ dotnet run

It'll be about 3-4 seconds. dotnet is the SDK and dotnet run will build and run your source code. Here's a short bit from the docs:

The dotnet run command provides a convenient option to run your application from the source code with one command. It's useful for fast iterative development from the command line. The command depends on the dotnet build command to build the code. Any requirements for the build, such as that the project must be restored first, apply to dotnet run as well.

While this is super convenient, it's not totally obvious that dotnet run isn't something you'd go to production with (especially Hello World Production, which is quite demanding! ;) ).

Dotnet Publish then Dotnet YOUR.DLL for Production

Instead, do a dotnet publish, note the compiled DLL created, then run "dotnet tst.dll."

For example:

C:\Users\scott\Desktop\tst> dotnet publish
Microsoft (R) Build Engine version 15.3 for .NET Core
Copyright (C) Microsoft Corporation. All rights reserved.

tst -> C:\Users\scott\Desktop\tst\bin\Debug\netcoreapp2.0\tst.dll
tst -> C:\Users\scott\Desktop\tst\bin\Debug\netcoreapp2.0\publish\
C:\Users\scott\Desktop\tst> dotnet run .\bin\Debug\netcoreapp2.0\tst.dll
Hello World!

On my machine, dotnet run is 2.7s, but dotnet tst.dll is 0.04s.

.NET Core is fast

Dotnet publish --self-contained

I could then publish a complete self-contained app - I'm using Windows, so I'll publish for Windows but you could even build on a Windows machine but target a Mac runtime, etc and that will make a \publish folder.

C:\Users\scott\Desktop\tst> dotnet publish  --self-contained -r win10-x64
Microsoft (R) Build Engine version 15.3 for .NET Core
Copyright (C) Microsoft Corporation. All rights reserved.

tst -> C:\Users\scott\Desktop\tst\bin\Debug\netcoreapp2.0\win10-x64\tst.dll
tst -> C:\Users\scott\Desktop\tst\bin\Debug\netcoreapp2.0\win10-x64\publish\
C:\Users\scott\Desktop\tst> .\bin\Debug\netcoreapp2.0\win10-x64\publish\tst.exe
Hello World!

Note in this case I have a "Self-Contained" app, so all of .NET Core is in that folder and below. Here I run tst.exe, not dotnet.exe because now I'm an end-user.

The results of a published .NET Core App

I hope this helps clear things up.

Sponsor: Check out JetBrains Rider: a new cross-platform .NET IDE. Edit, refactor, test, build and debug ASP.NET, .NET Framework, .NET Core, or Unity applications. Learn more and get access to early builds!

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

facebook twitter subscribe
About   Newsletter
Sponsored By
Hosting By
Dedicated Windows Server Hosting by SherWeb

Exploring CQRS within the Brighter .NET open source project

June 25, '17 Comments [15] Posted in DotNetCore | Open Source
Sponsored By

The logo for the "Brighter" Open Source project is a little cannon. Fire and Forget?There's a ton of cool new .NET Core open source projects lately, and I've very much enjoyed exploring this rapidly growing space. Today at lunch I was checking out a project called "Brighter." It's actually been around in the .NET space for many years and is in the process of moving to .NET Core for greater portability and performance.

Brighter is a ".NET Command Dispatcher, with Command Processor features for QoS (like Timeout, Retry, and Circuit Breaker), and support for Task Queues"

Whoa, that's a lot of cool and fancy words. What's it mean? The Brighter project is up on GitHub incudes a bunch of libraries and examples that you can pull in to support CQRS architectural styles in .NET. CQRS stands for Command Query Responsibility Segregation. As Martin Fowler says, "At its heart is the notion that you can use a different model to update information than the model you use to read information." The Query Model reads and the Command Model updates/validates. Greg Young gives the first example of CQRS here. If you are a visual learner, there's a video from late 2015 where Ian Cooper explains a lot of this a the London .NET User Group or an interview with Ian Cooper on Channel 9.

Brighter also supports "Distributed Task Queues" which you can use to improve performance when you're using a query or integrating with microservices.

When building distributed systems, Hello World is NOT the use case. BUT, it is a valid example in that it strips aside any business logic and shows you the basic structure and concepts.

Let's say there's a command you want to send. The GreetingCommand. A command can be any write or "do this" type command.

internal class GreetingCommand : Command
public GreetingCommand(string name)
:base(new Guid())
Name = name;

public string Name { get; private set; }

Now let's say that something else will "handle" these commands. This is the DoIt() method. No where do we call Handle() ourselves. Similar to dependency injection, we won't be in the business of calling Handle() ourselves; the underlying framework will abstract that away.

internal class GreetingCommandHandler : RequestHandler<GreetingCommand>
[RequestLogging(step: 1, timing: HandlerTiming.Before)]
public override GreetingCommand Handle(GreetingCommand command)
Console.WriteLine("Hello {0}", command.Name);
return base.Handle(command);

We then register a factory that takes types and returns handlers. In a real system you'd use IoC (Inversion of Control) dependency injection for this mapping as well.

Our Main() has a registry that we pass into a larger pipeline where we can set policy for processing commands. This pattern may feel familiar with "Builders" and "Handlers."

private static void Main(string[] args)
var registry = new SubscriberRegistry();
registry.Register<GreetingCommand, GreetingCommandHandler>();

var builder = CommandProcessorBuilder.With()
.Handlers(new HandlerConfiguration(
subscriberRegistry: registry,
handlerFactory: new SimpleHandlerFactory()
.RequestContextFactory(new InMemoryRequestContextFactory());

var commandProcessor = builder.Build();


Once we have a commandProcessor, we can Send commands to it easier and the work will get done. Again, how you ultimately make the commands is up to you.

commandProcessor.Send(new GreetingCommand("HanselCQRS"));

Methods within RequestHandlers can also have other behaviors associated with them, as in the case of "[RequestLogging] on the Handle() method above. You can add other stuff like Validation, Retries, or Circuit Breakers. The idea is that Brighter offers a pipeline of handlers that can all operate on a Command. The Celery Project is a similar project except written in Python. The Brighter project has stated they have lofty goals, intending to one day handle fault tolerance like Netflix's Hystrix project.

One of the nicest aspects to Brighter is that it's prescriptive but not heavy-handed. They say:

Brighter is intended to be a library not a framework, so it is consciously lightweight and divided into packages that allow you to consume only those facilities that you need in your project.

Moving beyond Hello World, there are more fleshed out examples like a TaskList with a UI, back end Http API, a Mailer service, and core library.

Be sure to explore Brighter's excellent documentation and examples, but be aware, this is a project under active development. Perhaps if you're new to OSS, if you find a broken link or two or a misspelling, you can do Your First Pull Request with a small fix?

Do be aware, again, that CQRS is not for every project. It's non-trivial and it's a "mental leap" as Martin Fowler puts it. If you buy in, you're adding complexity...for a reason. Keep your eyes open and do your research. It's a great pattern if you have a high performance/volume application that struggles with write concurrency or a flaky backend.

In fact there are quite a few mature CQRS libraries in the .NET open source space. I'll explore a few - which are your favorites?

Sponsor: Seq is simple centralized logging, on your infrastructure, with great support for ASP.NET Core and Serilog. Version 4 adds integrated dashboards and alerts - check it out!

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

facebook twitter subscribe
About   Newsletter
Sponsored By
Hosting By
Dedicated Windows Server Hosting by SherWeb

How to reference a .NET Core library in WinForms - Or, .NET Standard Explained

June 16, '17 Comments [13] Posted in DotNetCore
Sponsored By

I got an interesting email today. The author said "I have a problem consuming a .net core class library in a winforms project and can't seem to find a solution." This was interesting for a few reasons. First, it's solvable, second, it's common, and third, it's a good opportunity to clear a few things up with a good example.

To start, I emailed back with "precision questioning." I needed to assert my assumptions and get a few very specific details to make sure this was, in fact, possible. I said. "What library are you trying to use? What versions of each side (core and winforms)? What VS version?"

The answer was "I am working with VS2017. The class library is on NETCoreApp 1.1 and the app is a Winforms project on .NET Framework 4.6.2."

Cool! Let's solve it.

Referencing a .NET Core library from WinForms (running .NET Full Framework)

Before we parse this question. Let's level-set.

.NET is this big name. It's the name for the whole ecosystem, but it's overloaded in such a way that someone can say "I'm using .NET" and you only have a general idea of what that means. Are you using it on mobile? in docker? on windows?

Let's consider that ".NET" as a name is overloaded and note that there are a few "instances of .NET"

  • .NET (full) Framework - Ships with Windows. Runs ASP.NET, WPF, WinForms, and a TON of apps on Windows. Lots of businesses depend on it and have for a decade. Super powerful. Non-technical parent maybe downloads it if they want to run or a game.
  • .NET Core - Small, fast, open source, and cross-platform. Runs not only on Windows but also Mac and a dozen flavors of Linux.
  • Xamarin/Mono/Unity - The .NET that makes it possible to write apps in C# or F# and run them on everything from an iPad to cheap Android phone to a Nintendo Switch.

All of these runtimes are .NET. If you learn C# or F# or VB, you're a .NET Programmer. If you do a little research and google around you can write code for Windows, Mac, Linux, Xbox, Playstation, Raspberry Pi, Android, iOS, and on and on. You can run apps on Azure, GCP, AWS - anywhere.

What's .NET Standard?

.NET Standard isn't a runtime. It's not something you can install. It's not an "instance of .NET."  .NET Standard is an interface - a versioned list of APIs that you can call. Each newer version of .NET Standard adds more APIs but leaves older platforms/operating systems behind.

The runtimes then implement this standard. If someone comes out with a new .NET that runs on a device I've never heard of, BUT it "implements .NET Standard" then I just learned I can write code for it. I can even use my existing .NET Standard libraries. You can see the full spread of .NET Standard versions to supported runtimes in this table.

Now, you could target a runtime - a specific .NET - or you can be more flexible and target .NET Standard. Why lock yourself down to a single operating system or specific version of .NET? Why not target a list of APIs that are supported on a ton of platforms?

The person who emailed me wanted to "run a .NET Core Library on WinForms." Tease apart that statement. What they really want is to reuse code - a dll/library specifically.

When you make a new library in Visual Studio 2017 you get these choices. If you're making a brand new library that you might want to use in more than one place, you'll almost always want to choose .NET Standard.

.NET Standard isn't a runtime or a platform. It's not an operating system choice. .NET Standard is a bunch of APIs.

Pick .NET Standard

Next, check properties and decide what version of .NET Standard you need.

What version of .NET Standard?

The .NET Core docs are really quite good, and the API browser is awesome. You can find them at 

The API browser has all the .NET Standard APIs versioned. You can put the version in the URL if you like, or use this nice interface.

API Browser

You can check out .NET Standard 1.6, for example, and see all the namespaces and methods it supports. It works on Windows 10, .NET Framework 4.6.1 and more. If you need to make a library that works on Windows 8 or an older .NET Framework like 4.5, you'll need to choose a lower .NET Standard version. The table of supported platforms is here.

From the docs - When choosing a .NET Standard version, you should consider this trade-off:

  • The higher the version, the more APIs are available to you.
  • The lower the version, the more platforms implement it.

In general, we recommend you to target the lowest version of .NET Standard possible. The goal here is reuse. You can also check out the Portability Analyzer and run it on your existing libraries to see if the APIs you need are available.

.NET Portability Analyzer

.NET Standard is what you target for your libraries, and the apps that USE your library target a platform.

Diagram showing .NET Framework, Core, and Mono sitting on top the base of .NET Standard

I emailed them back briefly, "Try making the library netstandard instead."

They emailed back just a short email, "Yes! That did the trick!"

Sponsor: Big thanks to Raygun! Don't rely on your users to report the problems they experience. Automatically detect, diagnose and understand the root cause of errors, crashes and performance issues in your web and mobile apps. Learn more.

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

facebook twitter subscribe
About   Newsletter
Sponsored By
Hosting By
Dedicated Windows Server Hosting by SherWeb
Page 1 of 5 in the DotNetCore category Next Page

Disclaimer: The opinions expressed herein are my own personal opinions and do not represent my employer's view in any way.