@tall-librarian-49374 @wet-noon-14291 just pushed v2 and now it supports nested types! This week I'll also rename the repository so Tomas can push his Kubernetes stuff and I'll try then and add AWS, too

It looks really good, I just need to shorten the nested types, instead of "virtualMachineStorageOsDisk" it can just be "storageOsDisk" when it's nested...

this compiles and works ^

who is using pulumi dotnet on azure in production?

Hello, I had a query about using component resources arguments in Pulumi. More specifically how I can extend some of the existing resources args, rather than re-declaring them all in our own classes? For example in a use case of having a component resource for a Virtual network and subnets, I would like to expose the majority of the resource args into the component resource but add some additional arguments for conditional creation and loops or to set a default value for some of them. I can see all of the args classes are sealed, so is the standard to re-declare them in our own class or to create another that includes these args classes as properties?

I have a .net core web api that requires cosmos db, azure file share, and I was wondering how you guys structure the solution so you can deploy a .net core web app/api using azure-csharp. Would you create a new project or would you install pulumi in the api you want to deploy?

@wet-noon-14291 I've updated the repo as discussed, now it is generic and has Kubernetes and Aws together with Pulumi.FSharp.Azure. Btw, I've published the Pulumi.FSharp.Aws, haven't tested it, but it should work.

And now compilation is waaay faster... From 12 hours to, probably few seconds

That's the new URL to the repo https://github.com/UnoSD/Pulumi.FSharp.Extensions

I've also published

Pulumi.FSharp.Kubernetes

on NuGet... but that is mostly to reserve the name as I was getting a bunch of errors when generating. I think @wet-noon-14291 worked a bit on it so he'll probably have it already working.

I don't even have an AWS account, so I would appreciate any feedback on it as I can't really test it

Hello, I have a question regarding deploying Azure Functions. Is it possible to access the functions keys as output somehow?

It seems to be present in typescript but not in .NET (https://github.com/pulumi/pulumi-azure/pull/297/files/fd45274e6ebb9e1412020904f69e9b64c8ef093a)

While I'm looking for my workaround, can I execute azure cli commandos in Pulumi somehow?

Is there a way to tell if my azure function is deployed when using

WEBSITE_RUN_FROM_PACKAGE

and blob storage uri? I want to fetch my function keys but I'm getting errors as I'm suspecting the function app is not fully deployed yet.

Has anyone experience a 401 nuget error when running Pulumi on Azure Pipelines when your NuGet.config has an internal/private nuget source?

Quick question. Trying to figure out how I would specify multiple Path Patterns for the AWS ALB ListenerRule. From the docs it appears that Values can be a Liststring however, when you attempt to supply multiple values the IDE indicates that only one value can be supplied. Just kinda curious as I haven't found a solution yet, and haven't found one digging through any examples.

I have annoying scenario where by our Azure Application Gateway always shows a diff on RequestRoutingRules because the way the values are stored appear to be in a different order than when I generate them. Does anyone else have this problem? I did try to order the RequestRoutingRules by name, but I can't order by a value that's then inside the Input<string> property. Anyone got any ideas?

hi folks, I'm trying to invoke something AFTER a resource gets created. I'm using

resource.Property.Apply(x => do something with x)

, but "do something with x" gets invoked BEFORE the resource creation (I presume as the property is the name and it's available upfront). I tried adding other properties with

Output.Tuple

, but it seems that, as the name is ready it will be always invoked immediately. Is there a way to wait for the resource to be created to execute some other code? Thanks

I'm porting some code from typescript to c# and fighting the static typing models...

Was curious if anyone else is experiencing this issue or not. When creating and RDS Cluster in AWS, for now I statically supplied a list of CloudwatchLogsExports. With no changes it keeps flip flopping the order instead of staying true to the order I supplied. Since the order is flip flopping it also indicated a (replacement) of the ClusterInstances, even though that is an incorrect state change.

I have an ASP.NET application which has a Docker File and am trying to create the Docker Image using Pulumi. When I specify the Docker.ImageArgs, I can set the Build to the relative path of the directory where the Docker File is located. However, I need to pass the working directory to Docker. I am currently doing the following when using Docker build: docker build -f ./Dockerfile [/working_directory_path] When using Pulumi (C#), I am doing something like: var image = new Docker.Image("app-img", new Docker.ImageArgs { Build = "../path_to/directory_of_docker_file", ImageName = appRepo.RepositoryUrl, Registry = new Docker.ImageRegistry { Server = appRepo.RepositoryUrl, Username = appRepoCredentials.GetAt(0), Password = appRepoCredentials.GetAt(1) }, }); However I need to be able to specify the working directory for the docker file (in a similar way to when using docker build). Does anyone know how to achieve this with the C# API?

I am trying to create a Fargate service, without elastic load balancing. The aws-cs-fargate example had a load balancer and exposes the public ip of the service using the DnsName property of the load balancer. However, if you do not create a load balancer, how do you get the public ip for the service? The service is being created with NetworkConfiguration = new Ecs.Inputs.ServiceNetworkConfigurationArgs { AssignPublicIp = true, }

Has anyone else encountered this error when running Pulumi config set on Azure DevOps?

2020-08-18T21:11:57.9524504Z ##[error]Error: There was an error when attempting to execute the process 'D:\a\_temp\1db5fca2-f664-4b02-a788-4cbea994050f\pulumi\bin\pulumi.exe'. This may indicate the process failed to start. Error: spawn D:\a\_temp\1db5fca2-f664-4b02-a788-4cbea994050f\pulumi\bin\pulumi.exe ENOENT

I’m getting the following error when I try to execute pulumi up in azure devops pipelines (which is using the pulumi task with the az backend)

error: no stack named 'cde' found

However when I execute this locally (using my PULUMI_ACCESS_TOKEN) everything works fine. Any ideas?

I just got a this error when I was trying to deploy this stack to azure. 2020-08-20T203848.0594180Z

panic: runtime error: invalid memory address or nil pointer dereference

I’m trying to create an app using Pulumi.Okta, has anyone done that before?

I’m currently setting some config in my Azure Pipeline, I was wondering if you guys know a way to this in batches or in a different way.

- task: Pulumi@1
  displayName: 'Set azure:location'
  inputs:
    command: 'config'
    args: 'set azure:location ${{ parameters.azureLocation }}'
    cwd: '$(Pipeline.Workspace)/${{ parameters.pulumiArtifactName }}'
    stack: '${{ parameters.pulumiStack }}'
    createStack: true
    loginArgs: '--cloud-url ${{ parameters.pulumiCloudUrl }}'

- task: Pulumi@1
  displayName: 'Set azureServiceBusConnectionString'
  inputs:
    command: 'config'
    args: 'set azureServiceBusConnectionString $(azureServiceBusConnectionString)'
    cwd: '$(Pipeline.Workspace)/${{ parameters.pulumiArtifactName }}'
    stack: '${{ parameters.pulumiStack }}'
    createStack: true
    loginArgs: '--cloud-url ${{ parameters.pulumiCloudUrl }}'

- ${{ if eq(parameters.setMongoConnectionString, 'true') }}:
  - task: Pulumi@1
    displayName: 'Set mongoConnectionString'
    inputs:
      command: 'config'
      args: 'set mongoConnectionString $(mongoConnectionString)'
      cwd: '$(Pipeline.Workspace)/${{ parameters.pulumiArtifactName }}'
      stack: '${{ parameters.pulumiStack }}'
      createStack: true
      loginArgs: '--cloud-url ${{ parameters.pulumiCloudUrl }}'

After my deployment runs, I'd like to add a stored procedure and some data to the (Azure Document) database. Is there a way to: 1. tell that the deployment has been performed (rather than previewed or cancelled) 2. get the exported values to find the connection string

Hello, could would anyone be able to tell me how to properly package up a Maven build Java Lambda Function? Maven produces the *.JAR file which then if you were to do it manually the *.JAR gets uploaded to the function. Then your handler would be PATH.CLASS::Method. I have tried a variety of ways, but I cannot figure out if its my handler that is incorrect or whether it's how it is being packaged. Above is the code, and for simplicities sake, if I were to upload the *.JAR manually to AWS the handler is

test.app.controller.AppController::Handler