AzureDevopsLabs | A collection of labs demonstrating DevOps tools that can be | Azure library

First of all, yes you can publish your artifact without actually publishing them on the pipeline as artifact. However, if you publish them on the pipeline and have build and release pipelines separate you can redeploy without rebuilding your app.

Also you can always download artifact and publish it manually if where is a need for that. (some development/debug purposes for instance).

And Zip deploy method

which is used here

Creates a .zip deployment package of the chosen Package or folder and deploys the file contents to the wwwroot folder of the App Service name function app in Azure. This option overwrites all existing contents in the wwwroot folder. For more information, see Zip deployment for Azure Functions.

so if you need to deploy it from the build pipeline you can just put path to your zip file or to folder file where the content is.

And please consider multi stage pipeline where you can separate build and deploy phases among stages. However, in this case you need to publish your artifact as a pipeline artifact. And this pipeline artifact will be downloaded automatically on deployment job.

I think you are missing a task to publish the test results to Azure DevOps. You are executing dotnet test and get a trx- file (the result file), but you are not doing anything with it.

You need to use the Publish Test Result task (https://docs.microsoft.com/en-us/azure/devops/pipelines/tasks/test/publish-test-results?view=azure-devops&tabs=yaml) afterward to upload the trx- file to Azure DevOps.

Or you change the dotnet test command to use the Visual Studio Test task (https://docs.microsoft.com/en-us/azure/devops/pipelines/tasks/test/vstest?view=azure-devops). This tasks makes the upload of the trx- file automatically.

Read a similar question from VSTS (Visual Studio Team Services) and was able to solve it.

Created a Pipeline variable called ConnectionStrings.DefaultConnection that had a reference value to my linked variable group.

If my secret was named ConnectionStringsDefaultConnection I would hook this up as a linked variable and then add $(ConnectionStringsDefaultConnection) as a value.

Source:

https://stackoverflow.com/a/47787972/3850405

have seen in the documentation that it cannot be used in the same way as a normal (non-secret) variable. It says "Instead, we suggest that you map your secrets into environment variables." and "You need to explicitly map secret variables."

The doc is misunderstood, it's not saying the secret variables cannot be used in the same way as a normal variable, instead, it says it's not suggested to pass secrets on the command line directly since some operating systems log command line arguments which could cause information leakage as mentioned, and it's suggested to map your secrets into environment variables.

This blog says you can use it directly in the script, which seems to contradict the MS document.

As mentioned above, the secrets can be used directly in the script although it's not suggested, so they are not contradictory.

You can also check the example in the MSDN doc: