Publish a .NET Core Worker Service to Azure

In last week’s post, .NET Core Worker Service, we created a .NET Core Worker Service and then showed how to host it as a Windows Service. This week we will be taking the application created in last week’s post and publishing it to Azure. If you haven’t read last week’s post I recommend you at least get through the application creation bits. Feel free to skip the part about making the application run as a Windows Service since we will be pushing to Azure for this post.

Publish to Azure

This post is going to assume you already have an active Azure account. If not you can sign up for a free Azure account. This post is also going to be using Visual Studio 2019 Preview for the publication process.

In Visual Studio right-click on the project and click Publish.

Since this project already has a folder publish setup from last week’s post my screenshot might look different from yours. Click the New link.

This will launch the Publish profile dialog. We will be publishing to Azure WebJobs using the Create New option. After verifying the settings are correct click Create Profile.

The next screen gets the details of the App Service that will be created. I did a new resource group instead of using the one that was defaulted in and a new hosting plan so that I could utilize the free tier, but both of these changes are optional. When you have everything set click the Create button.

Once create is clicked it takes a couple of minutes to deploy the application. This first deploy is going to be a throwaway for us. There are a few settings we need to tweak to get the application running the way we want. After the initial deployment is complete we end up back on the profile details screen. Click the Edit link.

The Publish Profile Settings dialog will show. Since this was written while .NET Core 3 is still in preview I’m using the Self-contained option for Deployment Mode. If you are doing this after the final release then you can skip this step. The second thing I am changing is the WebJob Type and I’m setting it to Continuous. This is because our Service is controlling its own work schedule and not being triggered by something else. Click Save to commit the changes and return to the publish dialog.

Now push the Publish button to push the application to Azure.

Use Azure Portal to Verify Application is Working

Now that our application is running in Azure how do we verify that it is actually executing as we expect? Since the only thing the sample application does it output some logs we are going to have to find a way to show the output of the logging. To start head over to the Azure Portal. Since we just published the application the resource we are interested in should show in your Recent resources section of the Azure Portal homepage. As you can see in the screenshot the one we are interested in is the first option and matches the name we saw above in Visual Studio. Click on the resource.

From the menu section scroll down to the Monitoring section and click App Service logs. Now turn On the option for Application Logging (Filesystem) and for the Level option select Information since that is the log level our sample application is outputting. Then click Save.

Back in the same Monitoring group of the menu select Log stream.

Wrapping Up

Running a worker in Azure ended up being pretty simple. Not that I’m surprised Microsoft tends to make all their products work well with Azure. I’m sure you will find a way to make a more useful worker than we had in this example.


Also published on Medium.

4 thoughts on “Publish a .NET Core Worker Service to Azure”

  1. Hi thanks for the nice article. I have been trying to understand if there is any limits to worker service (net core) compared to webjob (net framework) for use in azure. Do you have any comments on this? I am looking for a continuous background worker type of service.

Leave a Reply to Erik Thysell Cancel Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.