app service envs (ase) are not automatable, so you have to first create it manually and add app service plans automatically afterwards

Thanks a lot for your help and all the very best

No worries! Feel free to ask about whatever else you need

For anyone else wanting to do this, there is currently a showstopper in the azure API. Azure App Service Plans cannot be automated in terraform and pulumi atm: https://github.com/Azure/azure-rest-api-specs/issues/3747

Hi everyone, is there any working example for @stocky-spoon-28903’s hint regarding async/await nodejs calls? I am trying to get it running with azure resource groups (

const resourceGroup = await azure.core.getResourceGroup …

) but am stuck in reading about the typescript Promise type and brain compiling to javascript. Any hints would be highly appreciated. As I am new to async/await that might be my main problem. @stocky-spoon-28903’s example seems to be just the core and I am just not getting how to do my async/await call so that nodejs does not fail with an

ReferenceError: resourceGroup is not defined

when using

resourceGroup.name

later in the code.

hi all. I'm looking at the pulumi/helm integration and I'm wondering, is there an example about taking an existing helm chart and modifying it somehow? one thing that's annoying is using a broken he;m chart and having to fork it to fix, wondering if pulumi could help there

is there a way to pass certain properties down the component hierarchy (from parent -> child) without passing the property explicitly to the childs constructor? I.e. think of something like Reacts Context API.

Use case: Wanna pass down the

namespace

to child components in k8s.

@glamorous-printer-66548 that’s a pretty interesting scenario… would you mind opening a bug ?

I have a relatively big k8s pulumi stack (~10 microservices, each with a deployment and a service). When I run

pulumi destroy

it seems to run the deletion sequentially for each single resource and it is therefore really slow. Is this intentional / known?

In which @lemon-spoon-91807 recursively infers a type for some unboxed value

lol

There’s some cool use of new TypeScript features there - but the upshot of it for Pulumi users is that in the upcoming release, it will be much easier when using the Pulumi TypeScript SDK to convert an Input into an Output. Particularly useful for folks creating components.

Is the multilanguage support still a story for Pulumi? It's looking easier and easier to just build an ecosystem around the typescript side. How are python, go and others ever going to keep up with that?

Like I was looking at what .net support would look like and it's just so much work to even get close to typescript parity.

is there a way to add folks to a pulumi project without them being a member of the corresponding github org?

Hey, just having a play with pulumi after coming from TF and was looking for a way to persist random numbers/strings between deploys (like generating a random password for a ServicePrincipal) - I found this (https://pulumi-community.slack.com/archives/C84L4E3N1/p1533922772000478) but was hoping for a baked in provider to handle this for me.

What is the best way to reach out and perform an AWS action in the midst of my Pulumi code?

I don't think so?

I tried to create a helm chart and added a bad provider. Pulumi is now stuck in this state even if I removed that code entirely:

error: could not find plugin for provider 'urn:pulumi:pulumi-stack-polyverse-dot-com::aws-ts-eks::pulumi:providers:aws::default'

hey folks, we have a few node.js apps which have their own package.json in the repository root, hence we cannot / don’t want to convolute the apps’ package.json with the package.json of the pulumi code. Hence I opted to move the pulumi code incl. configs into an

infra

subdirectory. However It’d like to keep the

Pulumi.yaml

in the repo root to be able to run pulumi CLI commands in the repo root. However it seems pulumi assumes the package.json to be in the same directory as the Pulumi.yaml and if that’s not the case it crashes with a bunch of errors. Is there a way to specify in the Pulumi.yaml a custom location of the package.json? Also is there some general docs / schema reference for the Pulumi.yaml ?

following use case: We have multiple k8s pulumi programs (think each for a different microservice A,B,C) each in different repositories. All pulumi programs shall deploy to a shared k8s namespace that is named after the environment (e.g. staging, dev1, dev_john_doe) etc. Now the question is where (in which pulumi program) we should create the namespace itself. Is it somehow possible to put the namespace resource definition redundantly in each program (A,B,C) but make pulumi not complain if the namespace already exists? I.e. if I spin up program A, and afterwards B, B shouldn’t complain about the existing namespace. In other words: I want some sort of idempotent / upsert-like / “ensure-existence” operation as part of pulumi. With plain

kubectl apply

of k8s manifests that’s possible. Thoughts?

Hey all, please can someone help me debug this issue I'm having. Basically I'm getting the message

error: expected string or JSON map; got resource.Computed

. The issue appears to be lines 118-121 where I am trying to use the outputs of two resources to create a new policy. The

getBucketPolicy

function thus would be the issue. (If I comment out the bit creating the policy the rest of it seems fine)

So if I create everything except the policy first, and then uncomment the policy bit it'll work. I'm assuming it's not happy with the dependency chaining?

(Side note: I love pretty much everything about pulumi so far)

Hi! For some reason the

user_data

that I'm supplying to my ec2 instances isn't getting run. When I log in with ssh and run the same command (python -m SimpleHTTPServer), everything works as expected. Is there any way to debug this situation?

quick question regarding the pulumi providers. I see when I build the pulumi-aws provider it creates a binary,

~/pulumi/bin/pulumi-resource-aws

Does each provider (ie, gcp for example) create these? I can't find them installed on my system anywhere, so not sure how it works. Does pulumi itself shell out to these providers in some way?

how can i move a stack from one org to another?

Hi. I'm new to Pulumi and considering using it to deploy and manage our clusters. Is there a way to see why an update failed? In my case, I am trying to deploy a container application using

cloud.Service

and I received this message after running

pulumi update

for the first time:

error: Plan apply failed: 1 error occurred:
    
    * creating urn:pulumi:basic-web-container-dev::basic-web-container::cloud:service:Service$aws:ecs/service:Service::basic-web: timeout while waiting for state to become 'true' (last state: 'false', timeout: 10m0s)