hello guys. quick question here.. I want to use pulumi to set up some stuff in spinnaker. I see there is no official spinnaker provider in pulumi, but there is some provider available for terraform: https://github.com/armory-io/terraform-provider-spinnaker giving it a quick look, do you think is gonna be possible to use the

pulumi-tf-provider-boilerplate

in order to wrap that terraform provider ? is it a simple procedure for someone like me who has literally 0 experience with go ?

crd2pulumi

is a killer, except for this little nuisance: https://github.com/pulumi/crd2pulumi/issues/19

And I keep on being bitten by the issues I filed myself: https://github.com/pulumi/pulumi/issues/5318 😂

I have a GCP

BackendConfig

CRD generated via

crd2pulumi

. But after creating an instance of it, I can’t refer to

crdInstance.metadata.name

. It errors with:

Property 'name' does not exist on type 'OutputInstance<ObjectMeta | undefined>'.

In the generated code, this

ObjectMeta

is defined as:

export type ObjectMeta = k8s.types.input.meta.v1.ObjectMeta;

Am I missing something here?

Hi everyone!  I have a question I was hoping someone could help me with.  I'm using crosswalk to create an ECS instance with ELB and so far it has made it so much easier.  However, I don't quite understand the documentation when it explains how to add advanced settings - in my case, a different health check endpoint.  Currently my target group seems to be created automatically due to my passing the listener into the 

portMappings

 attribute of my ECS service, but I can't see a way to override any of those defaults without fully creating a target group myself.  If I have to do that I will lose all the benefit of crosswalk in the first place. TL;DR: what's the correct way to change the heath check path of the auto-generated target group when passing an application load balancer listener into an ECS service? Thanks in advance! 🙂

I'm trying to migrate some resources between kubernetes clusters in the same stack as we're going to move and then kill the old cluster. I have the new cluster up and am installing all the same helm charts into it one at a time. I'm on the first Helm chart and am getting an error because it installs CRDs. Pulumi is saying the new CRDs have a duplicate resource URN. I don't think I can do anything about that as all I can do in the Pulumi code is affect the name of the chart. Anyone know a way around this?

I'm surprised Pulumi doesn't give them unique 7 digit extensions like most things behind the scenes

What causes an error of Resource monitor is terminating in RegisterResourceOutputs? I keep hitting this error on one of my stacks and not sure why. The other stacks seem to be functioning normally.

Hey folks, I'm attempting to recycle the username and password on a RDS cluster here, but when it does pulumi is attempting to create the db from scratch and then fails due to the DB already existing

is there anyway around this

Hello everyone 👋 New Pulumi customer here. Loving the toolkit so far.

[EDIT: Corrected timing - the link below is also correct] I'll be hosting an office hours today at 4PM PDT / 10 AM AEDT / 11PM GMT. Feel free to stop by! https://pulumi-community.slack.com/archives/C0110AVM89Z/p1616490643005300

How long does the Pulumi CLI cache new project(s), I was creating a new project - it failed with a permissions error (I'll ask about that separately). After I wiped out the directory, the CLI is still telling me there's already a project created with that name.

Hi! Yesterday I started using Pulumi, followed the workshop. I really like it so far! Planning on migrating all my Terraform plans to Pulumi.

Bumping into a problem with

Promise leak detected

. I have my subclass of ComponentResource for a certain k8s abstraction, containing 2 resources. I pass the

args

unchanged to the second resource. As a bug in my code, I forgot to pass

args.metadata

to the first of my contained resources, mainly to have the correct k8s namespace for this resource as well. But after passing the metadata, I get the

Promise leak detected

. Below is a summary of my component:

export class Service extends pulumi.ComponentResource {

    constructor(name: string, args: kubernetes.core.v1.ServiceArgs, opts?: pulumi.ComponentResourceOptions) {
      super("ng:kubernetes:core:Service", name, {}, opts);
  
      this.backendcfg = new backendconfig.cloud.v1.BackendConfig(
        `${name}-backendconfig`,
        {
          metadata: args.metadata,
          ...
        },
        {
          parent: this,
        }
      )
  
      this.service = new kubernetes.core.v1.Service(
        name,
        args,
        ...
      );
  
  }

Hey, what about importing resources created with custom resource providers? According to this GH issue: https://github.com/pulumi/pulumi/issues/3550#issuecomment-557216373 it's still a problem?

Hi all, I'm seeing issues with 403s while installing plugins (looks like S3 permissions on Pulumi's end). I've opened up an issue https://github.com/pulumi/pulumi/issues/6610.

Hi, we use Pulumi a lot at our organisation and feel quite useful and interesting. Now that we thought we could take advantage of Pulumi Kubernetes Operator and drive stacks backed by Pulumi self-managed backend hosted on Azure Blob storage. When we try to update an existing stack from a repo, we get *unable to open bucket azblob://*****: azureblob.OpenBucket: accountName is required* as we couldn't get AZURE_STORAGE_ACCOUNT and AZURE_STORAGE_KEY environment variables during stack update. Does anyone know how to make these env variables available before or when operator tries to select the existing stack and update the stack?

Hi all, I have a question, and I think it is an easy one, and that I am just being very dumb. I have tried a few different little projects. I am just using

pulumi login --local

. These two projects are in different folders, with different yaml files. Each had their own stack created, but when I say

pulumi stack ls

in one of the projects folders, I seem to get both stacks accross multiple projects. I noticed this only because I tried to name the stack in both projects "dev". Am I doing something wrong here, or is this a byproduct of using local login or something?

question, the https://www.pulumi.com/pricing/ plan comparasion, says that starter plan has "Up to 20 Project stacks" Does this means that a single project can have maximum 20 stacks, or that across all projects we can have maximum 20 stacks? Also on the free plan this is not specified, does the same limit applies or another?

Hello everybody, I have a question about pulumi output. In my project, I am deploying infrastructure on AWS. I want to use "address" output from RDS instance, and use that as environment variable for ecs container definition. However, it seems that I cannot extract string from address output.

I've tried to do something like this master.address.apply(lambda ad: "{ad}") but it's not working. I'am using Python

Can you guys help me?

Hey, I've run in to an issue where I can't see one of my stacks with

pulumi stack ls

unless

PULUMI_CONFIG_PASSPHRASE

is set, I'm using the s3 backend and the passphrase secrets provider. Output of

pulumi --logtostderr -v5 stack ls

below:

I0324 18:37:18.745036   38931 backend.go:859] error reading stack: staging (decrypting secret value: failed to decrypt: incorrect passphrase, please set PULUMI_CONFIG_PASSPHRASE to the correct passphrase) skipping
NAME         LAST UPDATE  RESOURCE COUNT
management*  3 weeks ago  32

Does anyone have any ideas? I can't

pulumi up

right now because there's a reference to the staging stack and the passphrase is different 😅

yeah, I came across this... your passphrases need to be the same for the stack references to work.

Helllo I have an error with GCP

error: gcp:container/cluster:Cluster resource 'cluster_infratotal' has a problem: Computed attribute cannot be set

Hello, I am really puzzled by the kubernetes provider. I am trying to deploy Traefik using the helm chart in Go. The namespace is not created automatically, so I am creating it and throwing a DependsOn ResourceOption. What I have noticed is that not all resources are created in the namespace: the Service isn’t. Could it be because they are using the new List kind?

# Source: traefik/templates/service.yaml
apiVersion: v1
kind: List
items:
  - apiVersion: v1
    kind: Service
    metadata:
      name: traefik
      labels:
        <http://app.kubernetes.io/name|app.kubernetes.io/name>: traefik
        <http://helm.sh/chart|helm.sh/chart>: traefik-9.18.0
        <http://app.kubernetes.io/managed-by|app.kubernetes.io/managed-by>: Helm
        <http://app.kubernetes.io/instance|app.kubernetes.io/instance>: traefik
      annotations:
    spec:
      type: LoadBalancer
      selector:
        <http://app.kubernetes.io/name|app.kubernetes.io/name>: traefik
        <http://app.kubernetes.io/instance|app.kubernetes.io/instance>: traefik
      ports:
      - port: 80
        name: web
        targetPort: "web"
        protocol: "TCP"
      - port: 443
        name: websecure
        targetPort: "websecure"
        protocol: "TCP"

Hello, I am not sure what happened but one of our stacks is gone. I was wondering if there's a way to recover it from the backups (We use S3 as backend for state) I've downloaded the file but the import command doesn't work

error: could not deserialize deployment: unexpected end of JSON input

Yet another interesting case 🙂