I'm guessing your pulumi definition isn't in the root of the checked out code? If so, use the

Working Directory

setting to point it to the correct spot.

Hi folks! We need to create an Azure SQL Database of type “Serverless” with Pulumi, but I can’t figure out how to do it. More info on that here: https://docs.microsoft.com/en-us/azure/sql-database/sql-database-serverless In ARM terms, a serverless SQL database looks like this:

{
  "type": "Microsoft.Sql/servers/databases",
  "apiVersion": "2019-06-01-preview",
  "name": "ideliverable-licensing/Licensing",
  "location": "northeurope",
  "sku": {
    "name": "GP_S_Gen5",
    "tier": "GeneralPurpose",
    "family": "Gen5",
    "capacity": 1
  },
  "kind": "v12.0,user,vcore,serverless",
  "properties": {
    "collation": "SQL_Latin1_General_CP1_CI_AS",
    "maxSizeBytes": 17179869184,
    "catalogCollation": "SQL_Latin1_General_CP1_CI_AS",
    "zoneRedundant": false,
    "readScale": "Disabled",
    "readReplicaCount": 0,
    "autoPauseDelay": 60,
    "storageAccountType": "GRS",
    "minCapacity": 0.5
  }
}

But a lot of these options are missing from the corresponding

Pulumi.Azure

classes. How do we specify the “Gen5" SKU, the “serverless” kind, the minimum/maximum number of vcores, and the auto-pause delay?

Pulumi.Azure.Sql.DatabaseArgs

don’t include any of these as far as I can tell. Thankful for any and all guidance!

Hi all, I get the following error when I try to use the Pulumi Task in the pipeline

I get the following error now, any help would be appreciated [7:10 AM] ##[error]Error: There was an error when attempting to execute the process 'd:\a\_temp\425ce36a-f50c-48af-99c5-5559867ad702\pulumi\bin\pulumi.exe'. This may indicate the process failed to start. Error: spawn d:\a\_temp\425ce36a-f50c-48af-99c5-5559867ad702\pulumi\bin\pulumi.exe ENOENT

Any ideas as to how I might solve this?

Gettings this stubborn error during

pulumi up

from my Azure Pipeline:

Diagnostics:
  azure:core:ResourceGroup (licensing-dev-rg):
    error: Error building account: Error getting authenticated object ID: Error listing Service Principals: autorest.DetailedError{Original:adal.tokenRefreshError{message:"adal: Refresh request failed. Status Code = '401'. Response body: {\"error\":\"invalid_client\",\"error_description\":\"AADSTS7000215: Invalid client secret is provided.\\r\\nTrace ID: d537fece-7902-4d0b-8696-debcc5b85900\\r\\nCorrelation ID: 62a70c31-1c5d-4de2-9503-b6c6ba7f555c\\r\\nTimestamp: 2020-01-12 20:48:44Z\",\"error_codes\":[7000215],\"timestamp\":\"2020-01-12 20:48:44Z\",\"trace_id\":\"d537fece-7902-4d0b-8696-debcc5b85900\",\"correlation_id\":\"62a70c31-1c5d-4de2-9503-b6c6ba7f555c\",\"error_uri\":\"<https://login.microsoftonline.com/error?code=7000215>\"}", resp:(*http.Response)(0xc00014c120)}, PackageType:"azure.BearerAuthorizer", Method:"WithAuthorization", StatusCode:401, Message:"Failed to refresh the Token for request to <https://graph.windows.net/***/servicePrincipals?%24filter=appId+eq+%27***%27&api-version=1.6>", ServiceError:[]uint8(nil), Response:(*http.Response)(0xc00014c120)}

Using the Pulumi Task Extension for Azure Pipeline. My task YAML looks like this:

- task: Pulumi@1
    displayName: Create/update Azure resources
    inputs:
      azureSubscription: 'IDeliverable Azure Subscription'
      command: 'up'
      args: '--skip-preview --yes'
      cwd: 'infra/'
      stack: 'dev'
    env:
      PULUMI_ACCESS_TOKEN: $(PulumiAccessToken)
  - script: |
      echo "##vso[task.setvariable variable=appName;isOutput=true]$(pulumi stack output appName)"
    displayName: Set stack outputs as variables
    name: stackVariables

The service connection

IDeliverable Azure Subscription

exists and validates fine. The same service connection is also used successfully from other non-Pulumi tasks. I triple-checked the client secrets, even generated new ones and updated and verified, but still Pulumi fails with this. Any ideas?

Guys , I have a K8s cluster configured via Pulumi running on Aks. Now I have to change some settings in the cluster like roleBasedAccessControl . The problem is when I do Pulumi preview , it’s telling me that is gonna replace my complete cluster and recreate . This is something that I don’t want because there are some applications running there . Is there a way to avoid this or what are the options in this case ? Any ideas ?

Hi,

I'm trying to create a azure function and assign it a service identity, then give it a blobcontributer role on my storage.

I can acheive the expected results when I run the up command locally

but in the pipeline it fails

Looks like I don't have some permission

could anyone please shed some light on this? How does it work for me through my console if I dont have permissions?

i changed the pulumi name of an NS record which led to the expected preview of create / delete - but it did the create first which succeeded despite there being a clash, then it deleted "the old one" which left the DNS without the NS record at all - yet the stack had still had the resource in it so a second 'pulumi up' didnt fix it. a 'pulumi refresh' noticed it was missing and then i could recreate it. is it worth filing an issue for? is it likely to be fixable?

how do i import an existing key in Azure Key Vault? i have attempted each of these with an error that it is an invalid URI

subscriptions/{subscriptionId}/resourceGroups/{rgName}/providers/Microsoft.KeyVault/vaults/{kvName}/keys
subscriptions/{subscriptionId}/resourceGroups/{rgName}/providers/Microsoft.KeyVault/vaults/{kvName}/keys/{keyName}
subscriptions/{subscriptionId}/resourceGroups/{rgName}/providers/Microsoft.KeyVault/vaults/{kvName}/keys/{keyName}/{keyVersion}

Does

azure.cosmosdb.SqlContainer

support the new autopilot throughput feature?

Hi All

I would like to use the name or id of the pull request as a suffix while naming my resources during deployment. What would be the best way to achieve this?

Many CI services dump a bunch of information into environment variables, check with whatever CI platform you're using but that's usually where you find it

The TF provider 1.39.0 supports specifying the minimum TLS version for an Azure Functions App: https://github.com/terraform-providers/terraform-provider-azurerm/pull/5074 However it’s not available in the

Pulumi.Azure.AppService.Inputs.FunctionAppSiteConfigArgs

class. Is Pulumi using an older version of the TF provider? How can I find out which provider version is being used?

anyone tried to enable azure devspaces on an aks cluster using pulumi? The kubeconfigraw in base64 is giving me a headache...

Anyone got any notes on Azure DevOps working with Azure storage account state backend. Added Env Variables, but still just getting a error like this

b

Pulumi does not seem to locate a resource group outside of the current stack's subscription. While trying to set up access policies for an existing KeyVault with Pulumi, I get an error that says

resource group not found

. despite that the keyVault id contains the full resource id (with the subscription)

I am trying to setup an Azure CosmosDB GremlinGraph as documented here: https://www.pulumi.com/docs/reference/pkg/nodejs/pulumi/azure/cosmosdb/#GremlinGraph However the TS property "GremlinGraph" does not exist. I am using Pulumi v1.8.1 My package.json:

{
    "name": "azure-typescript",
    "devDependencies": {
        "@types/node": "^8.0.0"
    },
    "dependencies": {
        "@pulumi/pulumi": "^1.0.0",
        "@pulumi/azure": "^1.0.0",
        "@pulumi/kubernetes": "^1.0.0",
        "@pulumi/azuread": "^1.0.0"
    }
}

Hey. I'm trying to output the host key for an Azure Function App after deployment, but I can't seem to get it to work. I've been referencing the documentation, which shows that the

getHostKeys()

method is available for the FunctionApp resource. I'm using the ArchiveFunctionApp resource, which doesn't specifically list the method, but I'm hoping that I can still access the keys somehow? • Documentation: https://www.pulumi.com/docs/reference/pkg/nodejs/pulumi/azure/appservice/#FunctionApp-getHostKeys

"use strict";
const pulumi = require("@pulumi/pulumi");
const azure = require("@pulumi/azure");

// Create Storage Account (and container) and App Service Plan.
/// Removed to shorted example.

// Create a Function App and deploy code to it.
const functionApp = new azure.appservice.ArchiveFunctionApp("archiveFunctionApp", {
	name: "my-function-app",
	resourceGroupName: resourceGroup.name,
	archive: new pulumi.asset.FileArchive("MyFunctionApp"),
	plan: appServicePlan,
	account: storageAccount,
	container: storageContainer,
	version: "~2",
	appSettings: {
		FUNCTIONS_WORKER_RUNTIME: "powershell"
	}
});

exports.FunctionAppHostKeys = functionApp.getHostKeys();

Hey. I'm attempting to control the name of the blob that is used with

azure.appservice.ArchiveFunctionApp

, but when I add the

zipBlob

property and remove the

archive

property, I'm met with an error -

Error: Deployment [archive] must be provided.

The docs describe

zipBlob

as "The blob containing all the code for this FunctionApp". If I leave the

archive

property in place, two blobs are uploaded to Storage - one is my

zipBlob

and the other is the randomly named blob created by the

ArchiveFunctionApp

. I'm guessing that there is a way to do this, but the docs seem pretty light on details (in fact they don't seem to mention

archive

at all), so I'm not too sure where to go from here? • Documentation: https://www.pulumi.com/docs/reference/pkg/nodejs/pulumi/azure/appservice/#ArchiveFunctionApp

const functionApp = new azure.appservice.ArchiveFunctionApp("archiveFunctionApp", {
	name: "my-function-app",
	resourceGroupName: resourceGroup.name,
	archive: new pulumi.asset.FileArchive("./MyArchive"),
	plan: appServicePlan,
	account: storageAccount,
	container: storageContainer,
	zipBlob: zipBlob,
	version: "~2",
	appSettings: {
		FUNCTIONS_WORKER_RUNTIME: "powershell"
	}
});

i upgraded to

@pulumi/pulumi

to 1.9.1 and

@pulumi/azure

to 1.11.0 and now get a lot of debug messages around registering resources and data sources for azure in the

Diagnostics

section of stdout every time i run

up

or

pre

. i am not running these commands with a

--debug

arg. is this expected behavior, or is it possible i have something mis-configured? (see image in thread)