The stack state json looks like this

"outputs": {
                        "privateSubnets": [
                            "subnet-0ef90f8900aa4e533",
                            "subnet-06e84b274ae517d39",
                            "subnet-01465b8f34f6490ee"
                        ],
                        "publicSubnets": [
                            "subnet-0c914a9aaec54e2a7",
                            "subnet-0893e2cfd9de09404",
                            "subnet-0216f9627f4fb20ef"
                        ],
                        "vpcId": "vpc-0f0e8a3ca9c7c8b41"
                    },

My intention is to count the number of private/public subnets. The data type of outputs is

map[string]interface{}

.

for _, res := range stack.Deployment.Resources {
		privateSubnets := res.Outputs["privateSubnets"].(interface{})
		log.Infoln(privateSubnets.([]string))
    }

I would like to understand the type assertion to read through the values in the interface. Below is the error I get

--- FAIL: TestInfrastructure (319.75s)
panic: interface conversion: interface {} is []interface {}, not []string [recovered]
        panic: interface conversion: interface {} is []interface {}, not []string

HI team, I need some help with the below issue:

➜   pulumi stack init test --secrets-provider="<gcpkms://projects/****/locations/global/keyRings/********/cryptoKeys/*****>"
Created stack 'test'
error: secrets (code=PermissionDenied): rpc error: code = PermissionDenied desc = Permission 'cloudkms.cryptoKeyVersions.useToEncrypt' denied on resource 'projects/prj-83923-s-orbit-8935/locations/global/keyRings/*****/cryptoKeys/******' (or it may not exist).
➜

I created the KMS ring and key manually in the console, and added my email in the permissions with KMS admin, and using the same email with gcloud auth on cmd line. ------------------ Not sure what is missing, but i really need some help here fixing this, thanks.

Is it possible to create a postgresql provider from the output of a aws rdsInstance? im struggling with this:

pgqlProvider := postgresql.Endpoint.ApplyT(func(x string) (*pgql.Provider, error) {
			host := strings.Split(x, ":")[0]
			port, err := strconv.Atoi(strings.Split(x, ":")[1])
			if err != nil {
				return nil, err
			}
			return pgql.NewProvider(ctx, "psql", &pgql.ProviderArgs{
				Database:  kratosDbName,
				Username:  dbUsername,
				Password:  password.Result,
				Superuser: pulumi.Bool(true),
				Host:      pulumi.String(host),
				Port:      <http://pulumi.Int|pulumi.Int>(port),
			})
		})

because this produces a

pulumi.Output

which

does not implement pulumi.ResourceOption (missing applyResourceOption method)

Hi, did anyone worked on mongodb-atlas provider in pulumi ? If yes, can someone please help with a working example for mongodbatlas.EncryptionAtRest , with GCP KMS ? this is in Python.

Good afternoon! I'm looking for some help setting up SECRETS in an AWS ECS Task definition using the form

[ { name: "SECRET", valueFrom: "<resource-arn>" }]

where

resource-arn

is the pulumi Output from an

aws.ssm.Parameter

that I defined earlier in the code.

Because the ssm.Parameter's

arn

value is a Pulumi output, it's not known/resolvable at the time that I'm trying to set up my ECS task def. As I'm working in typescript, I'm used to working with promises - but I'm not sure how I should change the structure of my code to indicate to Pulumi that the

arn

is an Output that will need to be resolved and passed to

valueFrom

.

I am guessing that it has something to do with the

stack

, but I honestly am not understanding the difference between

getOutput

,

requireOutput

, and

requireOutputValue

.

Or if those are even valid invocations at the point in the code that I'd have to use them.

Looking at it again... even more confusingly, the typescript typings suggest that it should work.

And yet, in the output of pulumi up, it says this:

"valueFrom":"Calling [toJSON] on an [Output<T>] is not supported.\n\nTo get the value of an Output as a JSON value or JSON string consider either:\n    1: o.apply(v => v.toJSON())\n    2: o.apply(v => JSON.stringify(v))\n\nSee <https://pulumi.io/help/outputs> for more details.\nThis function may throw in a future version of @pulumi/pulumi."}

(And I've tried both of those)

Looking at it a little further, trying to actually log the value of the

arn

in an

apply

call, I'm seeing the account-id portion of the ARN is either being obfuscated or masked as * - is that what is actually being passed into AWS and if so, then it would make sense why I'm getting an error about invalid characters in my ARN.

And as suspected, if I just hardcode the ARN, it works. Little but stuck on where to go debugging the Output from here.

Hi, can we have multiple python files in pulumi project? Instead of just one main.py?

https://github.com/pulumi/pulumi-aws/issues/2017 Is there currently a known workaround to this?

Needing to update the value of an AWS SSM Parameter causes this error during pulumi up.

Even if I somehow have to delete and recreate the stack? looking for anything that will work at this point.

Is there a way to rename the logical name of a resource? The one registered in Pulumi?

Folks, I'm looking for a way to move an existing GCP service into being managed by Pulumi without having to delete the service and let Pulumi (re) create it. The stack thinks it needs to create a new service for prod. When I tried it out with dev and stage, I deleted the old service and let Pulumi create it anew. But I'd rather not do that in prod. Is there a way to let Pulumi know something already exists and it should just take over managing and updating it?

Does using a self-hosted backend affect the concurrency of deploying resources? The stacks I have on app.pulumi.com seem to update all resources in parallel that can be, but my stacks with an S3 backend seem to only update one resource at a time. Is that expected?

Is there any way to avoid the

k8s.helm.v3.Chart

resource recreating/updating resources every time, even though nothing has changed? Eg. the

ALB-intgress-controller

deployed using the

Chart

resource recreates the tls-secret every time and updates the

ALB-webhooks

when nothing has changed in the actual pulumi declaration of the resource

Hello, I am using : https://www.pulumi.com/registry/packages/kubernetes-ingress-nginx/ but I can't find a way to associate a loadbalancer IP or LoadBalancer separate service. This one is always a new one generated while I was able to specify it in a loadBalancer service. Is there's a way to use it instead of ingress-controler creates its own ?

Hi Folks, did anyone worked on mongodb atlas provider ? I am looking for examples for end-to-end setup or any example that could help. Please let me know if anyone has done any work on mongodb atlas, as am finding it hard to write codes for lot of stuff, for eg the below. is to enable encryption at rest, but there is no proper example and am new to this so unable to understand, any help would be highly appreciated , thanks. https://www.pulumi.com/registry/packages/mongodbatlas/api-docs/encryptionatrest/

👋 Hi everyone!

I want to install Vector helm charts on my K8s pods, but I am not able to pass provider for my below code

vector = Chart('vector', ChartOpts(
    fetch_opts={'repo': '<https://helm.vector.dev>'},
    chart='vector'
))

How to pass provider in python?. I have a custom provider code

custom_provider = Provider(
    "inflation_provider", kubeconfig=kube_config
)

Has anyone had an issue with setting up configuration keys? I add a single config using

pulumi config set

and then I get mysterious errors liek this with random resources that have no changes:

error: could not validate provider configuration: 1 error occurred:
        * Invalid or unknown key

Hi team, I’m wondering what registering resource outputs for a pulumi component does. It’s not too clear to me from documentation whether this is cosmetic or necessary for a pulumi component to be functional. If we have a

ctx.RegisterResourceOutputs(<component>, resourceOutputsMap)

at the end of our component’s constructor but before the return, will this call ensure that all values in

resourceOutputsMap

resolve before proceeding? I don’t see anything visual during

pulumi preview

in

Outputs

when adding this call (just the usual exports). Any help in understanding this method would be appreciated, as I’m debating whether or not I should call it.

Hi, is there a way we can reference resources from one pulumi project to another? For eg, i have a pulumi folder for GCP network, with dev and stage stack. Now i want to create another folder for compute engine and reference the respective env from network pulumi project, how can i achieve this? Any sort of help is highly appreciated! Thanks