Does

pulumi stack init

no longer create a new yaml file?

I’m not sure I understand the typescript support. If I want to have a couple of custom options in tsconfig.json, I need to disable typescript support in the YAML file and then build the TS myself before running pulumi?

CDK Day! https://www.cdkday.com/

Hey, I am having some trouble deploying a compute@edge service on Fastly. The services requires a dictionary with two items. In order to add items to a dictionary i need a dictionaryID i have the following code to get this id.

const dict = service.dictionaries.apply(d => d?.[0].dictionaryId || '1234') // <--- This always returns undefined, so default to 1234

however because of the comment above whenever i run

pulumi up

i get the following error.

Diagnostics:
  fastly:index:ServiceDictionaryItems (CredentialsDict):
    error: 1 error occurred:
    	* Error creating dictionary items: service 6JGsFBzFbONzfbcXDyfRP3, dictionary 1234, 400 - Bad Request:

        Title:  Bad request
        Detail: Couldn't find Dictionary '{ dictionary_id => 1234, service_id => 6JGsFBzFbONzfbcXDyfRP3 }'

  pulumi:pulumi:Stack (fastly-shell-dev):
    error: update failed

Is it possible to retrieve this ID during the deployment process? If not how can I query this after the service is deployed and still able to deploy the dictionary at the same time?

Curious to hear about how people solve release management with Pulumi. We have Cloud Run instances and need to update infrastructure independently from the application container version. How do you solve this ? Was thinking of having “ignoreChanges” on Cloud Run resources in Pulumi and then use our release management software to update Cloud Run images directly. But I wonder if there’s a way to do this entirely in Pulumi by targeting the CloudRun resources ?

I’m creating an ECS task definition with Typescript. This takes in a string for the container definitions. I want to set some environment variables in said container definitions that are secret, but when

config.requireSecret('something')

I get an output…which won’t be useable in the string. Is there a better way to get these values that would be different per stack?

Hello guys do we have a channel to discuss pulumi policy topics? thanks

Hello, I'm attempting to use stack references with Typescript and I'm having trouble figuring out the best way to get the stack output reference as the correct type. When I do a

requireOutput()

from the stack reference it always ends up as

Output<any>

type. Is there a way to set the expected type of the output so I can get something like

pulumi.Output<string>

or other types?

Hello guys. During CDK Day, somebody from Pulumi shared a demo about some “self service” tooling that displays a nice UI, etc. Can somebody point me to that information / project or person? Thanks.

Ack. I just accidentally overwrote a stack file,

Pulumi.prod.yaml

, before I had a chance to commit it. Is there any way of getting it back?

Hi. Does someone know how to setup Azure SQL Database in Server less mode with Pulumi C#?

Anyone know why Function Serialization doesn’t include node_modules packages in the uploaded code?

Hi all, trying to diagnose an issue in production with our Pulumi pipeline. It uses and AzureAD Provider (Terraform under the covers?) with a clientid/secret as follows:

var userTenantProvider = new Provider("UserTenantProvider", new ProviderArgs

{

TenantId = _config.Environment.ActiveDirectory.UserTenantId,

ClientId = _userTenantProviderConfig.ClientId,

ClientSecret = _userTenantProviderConfig.ClientSecret

});

The secret has been rotated and the new one is passed into the pipeline. However the pipeline still fails with the following message: _building client: unable to obtain access token: clientCredentialsToken: received HTTP status 401 with response: {"error":"invalid_client","error_description":"AADSTS7000215: Invalid client secret provided. Ensure the secret being sent in the request is the client secret value, not the client secret ID, for a secret added to app '[secret]'._ The pipeline was working fine before the secret changed, it's as if the old one is cached somewhere. Any ideas?

This message was deleted.

Is it possible to do tf2pulumi with OCI terraform code ?

Does Pulumi's AWS Crosswalk API Gateway generate Swagger? For example, can it generate a JSON if I type in: https://<some aws url>/Stage/swagger/swagger.json?

Hi, has anyone got

RetainOnDelete

to work with a Helm v3.Chart? If I delete a chart it continues to delete the associated k8s resources, instead of leaving them unchanged. Namespaces (for example) work as expected, so maybe it is something to do with Helm charts and/or the Go implementation? (Example code in the comments)

Hi all, I am new to Pulumi and just wanted to check whether its possible to generate the ARM or CloudFormationTemplate from Pulumi code?

Hi. I am an avid Pulumi enthusiast. I want to share a project I have been working on that I hope might find someone's interest here. It is cost estimations for Pulumi, such that you always know what your infrastructure code will cost before you deploy it to the cloud. I have built a prototype that works with a couple Azure resources, but it is still under development. If you are interested you can find out more here and sign up to my email list so I know that you find this useful, since it takes a lot of time to develop this : https://cloudcostify.app/ Thanks for your time.

Is there an integration to manage slack apps (webhooks, etc)? I don’t see one (and the regular slack integration doesn’t have it) but figured I’d ask in case there was something in development.

Hello all! New to Pulumi here, I'm currently looking for a solution that imports all of our AWS infrastructure to IaC. Tried Terracognita and Terraformer as well but Pulumi seems to be the right tool for the job. Will the "pulumi import" tool be able to import the entire infrastructure or will it have to be 1 by 1?

Hi all, I have an ApplicationLoadBalancer (awsx.elasticloadbalancingv2.ApplicationLoadBalancer) that I create in my base network stack that listens to all subdomain requests (*.domain.com), in a higher level application stack I want to add a ListenerRule that points to a ApplicationTargetGroup using the ApplicationLoadBalancer but I can't seem to find a way in doing so using the TypeScript SDK. Can anyone point me in the right direction? Thanks. Here is a pseudo code of what i've tried so far and the error I get: *Base network stack

const alb = new awsx.elasticloadbalancingv2.ApplicationLoadBalancer ........
const appListenerHttps = alb.createListener.......
const zone  = aws.route53.getZoneOutput({name:domainAddress})
const albDomain = new aws.route53.Record(`domain-record`, {
    name: pulumi.interpolate `*.${domainAddress}`,
    zoneId: zone.zoneId,
    type: 'CNAME',
    records: [appListenerHttps.endpoint.hostname]
  });
export const albListenerArn = appListenerHttps.listener.arn
export const albArn = alb.loadBalancer.arn

*Application stack

const networkingStack = new StackReference(config.require('stack.networking'));
const alb =   aws.applicationloadbalancing.getLoadBalancerOutput({arn:networkingStack.getOutput('albArn')});
const targetGroup = new awsx.elasticloadbalancingv2.ApplicationTargetGroup(
      `app-alb-tg`,
      {
        ...............
        loadBalancer: new awsx.elasticloadbalancingv2.ApplicationLoadBalancer('alb',{
           loadBalancer:alb <----error: Type 'GetLoadBalancerResult' is missing the following properties from type 'LoadBalancer': enableCrossZoneLoadBalancing, namePrefix, tagsAll, urn, getProvider
        })
      }
    );

const itoolListener = new aws.lb.ListenerRule(
      `app-alb-lstn`,
      {
        listenerArn: networkingStack.getOutput('albListenerArn'),
        priority: 100,
        actions: [
          {
            type: 'forward',
            targetGroupArn: targetGroup.targetGroup.arn
          }
        ],
        conditions: [
          {
            hostHeader: {
              values: [pulumi.interpolate`test.${domainAddress}`]
            }
          }
        ]
      }
    );
.....

Hi All, does pulumi supports a native way to apply some resources to a specific stack only?. Say I have 3 VM instances that I am preparing to be rolled out on staging stack only, and not yet ready to roll to prod stack, how can I enforce Pulumi to ignore these resources if someone do

pulumi up

on prod stack?

Hello, I am using Pulumi and Localstack to run a stack. I'm running into a problem with an ECS cluster. I start it using this code and it starts succesfully.

const ecsCluster = new aws.ecs.Cluster(
    'test',
    {
      name: 'test',
    },
    {
      provider,
    },
  );

awslocal ecs describe-clusters --cluster test --region eu-central-1

shows this

{
  "clusters": [
    {
      "clusterArn": "arn:aws:ecs:eu-central-1:000000000000:cluster/test",
      "clusterName": "test",
      "status": "ACTIVE",
      "registeredContainerInstancesCount": 0,
      "runningTasksCount": 0,
      "pendingTasksCount": 0,
      "activeServicesCount": 0
    }
  ],
  "failures": []
}

But when I do

pulumilocal down

I get this error

error: deleting urn:pulumi:localstack::Development::aws:ecs/cluster:Cluster::test: 1 error occurred:
    	* error waiting for ECS Cluster (arn:aws:ecs:eu-central-1:000000000000:cluster/test) to become Deleted: couldn't find resource (21 retries)

Even though it does exist. It does however get deleted properly

hi, i'm trying to set up a FargateService with a non-default VPC and I keep running into this error on preview

Diagnostics:
  pulumi:pulumi:Stack (toronto):
    error: Error: invocation of aws:ec2/getVpc:getVpc returned an error: invoking aws:ec2/getVpc:getVpc: 1 error occurred:
        * no matching EC2 VPC found
    
    
        at Object.callback (/home/julian/betx/infra/node_modules/@pulumi/runtime/invoke.ts:162:33)
        at Object.onReceiveStatus (/home/julian/betx/infra/node_modules/@grpc/grpc-js/src/client.ts:338:26)
        at Object.onReceiveStatus (/home/julian/betx/infra/node_modules/@grpc/grpc-js/src/client-interceptors.ts:426:34)
        at Object.onReceiveStatus (/home/julian/betx//infra/node_modules/@grpc/grpc-js/src/client-interceptors.ts:389:48)
        at /home/julian/betx/infra/node_modules/@grpc/grpc-js/src/call-stream.ts:276:24
        at processTicksAndRejections (node:internal/process/task_queues:78:11)

i am using a custom VPC for an ECS service, but trying to do

pulumi up -v 4...7

doesn't seem to show anything useful. how can I debug this? for reference, i deleted the default VPC.

Still seems like there’s a good opportunity to abstract away IAM policies in Pulumi, e.g. define a link between a lambda and another lambda, or an S3 bucket, and immediately get generated for you: • sensible and customizable IAM policies (e.g. invokeLambda, getObject) • environment variables in the lambda (ARN of other lambda or s3 bucket name)

Hi. Does anyone know why one needs to define two environment variables to support the

google native

provider and

Google Cloud Key Management Service (KMS)

provider? Pulumi uses:

GOOGLE_CREDENTIALS

to support:

Pulumi in a non-interactive setting (for example CI/CD systems, where a person can not complete the normal gcloud auth application-default login flow).

which I have configured since Day 1. However, I recently changed secrets provider to Google's KMS service and I now must also define the

GOOGLE_APPLICATION_CREDENTIALS

envvar with the same creds..? Why the need for 2 x envvars?

Note: my CICD env does not utilise the gcloud sdk. I am purposefully constructing a GitOps workflow where the gcloud sdk is not a dependency.

Is it not possible to run pulumi functions in a pulumi-policy? Do I have to use the normal aws-sdk? When trying i get `Error: Program run without the Pulumi engine available; re-run using the

pulumi

CLI`

Hi, does this work yet? I upgraded to eks provider 0.40 but it doesn't seem to fix it https://pulumi-community.slack.com/archives/C84L4E3N1/p1652470392669989