Developing .NET applications in RHEL 8

Red Hat Enterprise Linux 8

Installing and running .NET Core 3.1 for developing .NET applications in Red Hat Enterprise Linux 8

Red Hat Customer Content Services


.NET Core is a general purpose development platform featuring automatic memory management and modern programming languages. It allows users to build high-quality applications efficiently. .NET Core is available in Red Hat Enterprise Linux and OpenShift Container Platform via certified containers.
.NET Core offers the following features:
  • The ability to follow a microservices-based approach, where some components are built with .NET and others with Java, but all can run on a common, supported platform in Red Hat Enterprise Linux and OpenShift Container Platform.
  • The capacity to more easily develop new .NET Core workloads on Microsoft Windows. Customers can deploy and run on either Red Hat Enterprise Linux or Windows Server.
  • A heterogeneous data center, where the underlying infrastructure is capable of running .NET applications without having to rely solely on Windows Server.
.NET Core 3.1 is supported on Red Hat Enterprise Linux 7, Red Hat Enterprise Linux 8, and OpenShift Container Platform versions 3.3 and later.


This guide describes how to install .NET Core 3.1 on Red Hat Enterprise Linux 8 (RHEL). See the Red Hat Enterprise Linux documentation for more information about RHEL 8.

Chapter 1. Using .NET Core 3.1 on Red Hat Enterprise Linux

1.1. Installing .NET Core

This procedure installs the .NET Core 3.1 runtime with the latest 3.1 SDK. When a newer SDK becomes available, it automatically installs as a package update.


.NET Core 3.1 is included in the AppStream repositories for RHEL 8. The AppStream repositories are enabled by default on RHEL 8 systems.

  1. Install .NET Core 3.1 and all of its dependencies:

    $ sudo yum install dotnet-sdk-3.1 -y
  2. Run the following command to verify the installation:

    $ dotnet --info

1.2. Creating an application


  1. Create a new Console application in a directory called hello-world:

    $ dotnet new console -o hello-world
      The template "Console Application" was created successfully.
      Processing post-creation actions...
      Running 'dotnet restore' on hello-world/hello-world.csproj...
      Restore completed in 87.21 ms for /home/<USER>/hello-world/hello-world.csproj.
      Restore succeeded.
  2. Run the project:

    $ cd hello-world
    $ dotnet run
    Hello World!

1.3. Publishing applications

.NET Core 3.1 applications can be published to use a shared system-wide version of .NET Core or to include .NET Core. These two deployment types are called framework-dependent deployment (FDD) and self-contained deployment (SCD), respectively.

For Red Hat Enterprise Linux, we recommend publishing by FDD. This method ensures the application is using an up-to-date version of .NET Core, built by Red Hat, that includes a specific set of native dependencies. On the other hand, SCD uses a runtime built by Microsoft.


  1. Use the following command to publish a framework-dependent application:

    $ dotnet publish -f netcoreapp3.1 -c Release
  2. Optional: If the application is only for RHEL, trim out the dependencies needed for other platforms with these commands:

    $ dotnet restore -r rhel.8-x64
    $ dotnet publish -f netcoreapp3.1 -c Release -r rhel.8-x64 --self-contained false

1.4. Running applications in containers

This section shows how to use the ubi8/dotnet-31-runtime image to run a precompiled application inside a container.


  1. Create a new MVC project in a directory named mvc_runtime_example:

    $ dotnet new mvc -o mvc_runtime_example
    $ cd mvc_runtime_example
  2. Publish the project:

    $ dotnet publish -f netcoreapp3.1 -c Release
  3. Create the Dockerfile:

    $ cat > Dockerfile <<EOF
    ADD bin/Release/netcoreapp3.1/publish/ .
    CMD ["dotnet", "mvc_runtime_example.dll"]
  4. Build your image:

    $ podman build -t dotnet-31-runtime-example .
  5. Run your image.

    $ podman run -d -p8080:8080 dotnet-31-runtime-example
  6. View the result in a browser:

Chapter 2. Using .NET Core 3.1 on Red Hat OpenShift Container Platform

2.1. Installing image streams

.NET Core image streams are installed using image stream definitions from s2i-dotnetcore with the OpenShift client binary (oc). A script is available to facilitate removing, installing, and updating the image streams.

.NET Core image streams can be defined in the global openshift namespace or locally in a project namespace. Sufficient permissions are required to update the openshift namespace definitions.

2.1.1. Installing image streams using oc



  1. List the available .NET Core image streams:

    $ oc describe is dotnet [-n <namespace>]

    The output shows installed images or the message Error from server (NotFound) if no images are installed.

  2. Install the .NET Core image streams:

    $ oc create -f
  3. Include newer versions of existing .NET Core image streams:

    $ oc replace -f

2.1.2. Installing image streams using a script

The script can be used to install, remove, and update .NET Core image streams. Linux/macOS


  1. Download the script.
  2. Log in to the OpenShift cluster using the oc login command.
  3. Install or update the image streams:

    ./ --os rhel8 [--namespace <namespace>]  [--user <subscription_user> --password <subscription_password>]

The pull secret can be added by providing the --user and --password arguments. If a pull secret is already present, these arguments are ignored.

Run ./ --help for more information on using this script. Windows


  1. Download the script.
  2. Log in to the OpenShift cluster using the oc login command.
  3. Install or update the image streams:

    ./ --OS rhel8 [--Namespace <namespace>]  [-User <subscription_user> -Password <subscription_password>]

The PowerShell ExecutionPolicy may prohibit executing this script. To relax the policy, run Set-ExecutionPolicy -Scope Process -ExecutionPolicy Bypass -Force.

The pull secret can be added by providing the -User and -Password arguments. If a pull secret is already present, these arguments are ignored.

Run Get-Help .\install-imagestreams.ps1 for more information on using this script.

2.2. Deploying applications

2.2.1. Deploying applications from source


  1. Run the following commands to deploy the ASP.NET Core application, which is in the app folder on the dotnetcore-3.1 branch of the redhat-developer/s2i-dotnetcore-ex GitHub repository:

    $ oc new-app --name=exampleapp 'dotnet:3.1~' --build-env DOTNET_STARTUP_PROJECT=app
  2. Use the oc logs command to track progress of the build:

    $ oc logs -f bc/exampleapp
  3. View the deployed application once the build is finished:

    $ oc logs -f dc/exampleapp
  4. The application is now accessible within the project. To make the project accessible externally, use the oc expose command. You can then use oc get routes to find the URL:

    $ oc expose svc/exampleapp
    $ oc get routes

2.2.2. Deploying applications from binary artifacts

The .NET Core Source-to-Image (S2I) builder image can be used to build an application using binary artifacts that you provide.


  1. Publish your application as described in Section 1.3, “Publishing applications”. For example, the following commands create a new web application and publish it:

    $ dotnet new web -o webapp
    $ cd webapp
    $ dotnet publish -c Release
  2. Create a new binary build using the oc new-build command:

    $ oc new-build --name=mywebapp dotnet:3.1 --binary=true
  3. Start a build using the oc start-build command, specifying the path to the binary artifacts on your local machine:

    $ oc start-build mywebapp --from-dir=bin/Release/netcoreapp3.1/publish
  4. Create a new application using the oc new-app command:

    $ oc new-app mywebapp

2.3. Environment variables

The .NET Core images support a number of environment variables to control the build behavior of your .NET Core application. These variables can be set as part of the build configuration, or they can be added to the .s2i/environment file in the application source code repository.

Variable NameDescriptionDefault


Selects the project to run. This must be a project file (for example, csproj or fsproj) or a folder containing a single project file.



Selects the assembly to run. This must not include the .dll extension. Set this to the output assembly name specified in csproj (PropertyGroup/AssemblyName).

The name of the csproj file


When set to true, the application will be compiled ahead-of-time. This reduces startup time by reducing the amount of work the JIT needs to do when the application is loading.



Specifies the space-separated list of NuGet package sources used during the restore operation. This overrides all of the sources specified in the NuGet.config file. This variable cannot be combined with DOTNET_RESTORE_CONFIGFILE.



Specifies a NuGet.Config file to be used for restore operations. This variable cannot be combined with DOTNET_RESTORE_SOURCES.



Specifies a list of .NET tools to install before building the app. It is possible to install a specific version by post pending the package name with @<version>.



Specifies a list of NPM packages to install before building the application.



Specifies the list of test projects to test. This must be project files or folders containing a single project file. dotnet test is invoked for each item.



Runs the application in Debug or Release mode. This value should be either Release or Debug.



Specifies the verbosity of the dotnet build commands. When set, the environment variables are printed at the start of the build. This variable can be set to one of the msbuild verbosity values (q[uiet], m[inimal], n[ormal], d[etailed], and diag[nostic]).



Configures the HTTP or HTTPS proxy used when building and running the application, respectively.



When set to true, the source code will not be included in the image.



Used to specify a list of folders or files with additional SSL certificates to trust. The certificates are trusted by each process that runs during the build and all processes that run in the image after the build (including the application that was built). The items can be absolute paths (starting with /) or paths in the source repository (for example, certificates).



Uses a custom NPM registry mirror to download packages during the build process.



This variable is set to http://*:8080 to configure ASP.NET Core to use the port exposed by the image. Changing this is not recommended.



When set to true, disables restoring multiple projects in parallel. This reduces restore timeout errors when the build container is running with low CPU limits.



When set to true, the NuGet packages will be kept so they can be re-used for an incremental build.



When set to true, creates a tar.gz file at /opt/app-root/app.tar.gz that contains the published application.


2.4. Sample applications

Two sample applications are available for use with the .NET Core S2I builder.

2.4.1. Creating the MVC sample application

s2i-dotnetcore-ex is the default .NET Core Model, View, Controller (MVC) template application.

This application is used as the example application by the .NET Core S2I image and can be created directly from the OpenShift UI using the Try Example link.

The application can also be created with the OpenShift client binary (oc).


To create the sample application using oc:

  1. Add the .NET Core application:

    $ oc new-app dotnet:3.1~ --context-dir=app
  2. Make the .NET Core application externally accessible and show the URL:

    $ oc expose service s2i-dotnetcore-ex
    $ oc get route s2i-dotnetcore-ex

Additional resources

2.4.2. Creating the CRUD sample application

s2i-dotnetcore-persistent-ex is a simple Create, Read, Update, Delete (CRUD) .NET Core web application that stores data in a PostgreSQL database.


The application can be created using the OpenShift client oc as follows:

  1. Add the database:

    $ oc new-app postgresql-ephemeral
  2. Add the .NET Core application:

    $ oc new-app dotnet:3.1~ --context-dir app
  3. Add environment variables from the postgresql secret and database service name environment variable:

    $ oc set env dc/s2i-dotnetcore-persistent-ex --from=secret/postgresql -e database-service=postgresql
  4. Make the .NET Core application externally accessible and show the URL:

    $ oc expose service s2i-dotnetcore-persistent-ex
    $ oc get route s2i-dotnetcore-persistent-ex

Additional resources

Chapter 3. Migration to .NET Core 3.1

This chapter provides migration information for .NET Core 3.1.

3.1. Migration from previous versions of .NET Core

See the following Microsoft articles to migrate from previous versions of .NET Core to newer versions of .NET Core:

3.2. Migration from .NET Framework to .NET Core 3.1

Review the following information to migrate from the .NET Framework.

3.2.1. Migration considerations

Several technologies and APIs present in the .NET Framework are not available in .NET Core. If your application or library requires these APIs, consider finding alternatives or continue using the .NET Framework. .NET Core does not support the following technologies and APIs:

  • Windows Communication Foundation (WCF) servers (WCF clients are supported)
  • .NET remoting

Additionally, a number of .NET APIs can only be used in Microsoft Windows environments. The following list shows a few examples of these Windows-specific APIs:

  • Microsoft.Win32.Registry
  • System.AppDomains
  • System.Security.Principal.Windows

Consider using the .NET Portability Analyzer to identify API gaps and potential replacements. For example, enter the following command to find out how much of the API used by your .NET Framework 4.6 application is supported by .NET Core 2.1:

$ dotnet /path/to/ApiPort.dll analyze -f . -r html --target '.NET Framework,Version=4.6' --target '.NET Core,Version=2.1'

Several APIs that are not supported in the default version of .NET Core may be available from the Microsoft.Windows.Compatibility NuGet package. Be careful when using this NuGet package. Some of the APIs provided (such as Microsoft.Win32.Registry) only work on Windows, making your application incompatible with Red Hat Enterprise Linux.

3.2.2. .NET Framework migration articles

Refer to the following Microsoft articles when migrating from .NET Framework.

Legal Notice

Copyright © 2020 Red Hat, Inc.
The text of and illustrations in this document are licensed by Red Hat under a Creative Commons Attribution–Share Alike 3.0 Unported license ("CC-BY-SA"). An explanation of CC-BY-SA is available at In accordance with CC-BY-SA, if you distribute this document or an adaptation of it, you must provide the URL for the original version.
Red Hat, as the licensor of this document, waives the right to enforce, and agrees not to assert, Section 4d of CC-BY-SA to the fullest extent permitted by applicable law.
Red Hat, Red Hat Enterprise Linux, the Shadowman logo, the Red Hat logo, JBoss, OpenShift, Fedora, the Infinity logo, and RHCE are trademarks of Red Hat, Inc., registered in the United States and other countries.
Linux® is the registered trademark of Linus Torvalds in the United States and other countries.
Java® is a registered trademark of Oracle and/or its affiliates.
XFS® is a trademark of Silicon Graphics International Corp. or its subsidiaries in the United States and/or other countries.
MySQL® is a registered trademark of MySQL AB in the United States, the European Union and other countries.
Node.js® is an official trademark of Joyent. Red Hat is not formally related to or endorsed by the official Joyent Node.js open source or commercial project.
The OpenStack® Word Mark and OpenStack logo are either registered trademarks/service marks or trademarks/service marks of the OpenStack Foundation, in the United States and other countries and are used with the OpenStack Foundation's permission. We are not affiliated with, endorsed or sponsored by the OpenStack Foundation, or the OpenStack community.
All other trademarks are the property of their respective owners.