It looks like Pulumi does not support deploying Docker containers to Google App Engine Flex runtimes. I dug a little deeper, and this seems like an upstream problem. There is an issue in the Terraform and Google Cloud repos to add support for this runtime into configuration settings: https://github.com/terraform-providers/terraform-provider-google/issues/5025 https://github.com/GoogleCloudPlatform/magic-modules/pull/3110 Is this correct? I just want to check to make sure, because if so that means that the only options for deploying containerized apps on GCP at the moment with Pulumi are through Cloud Run (which is great but does not support websocket functionality so is a no-go for apps that need live subscriptions) or as a k8s pod, which can be heavy for a one-off service or two.

We are building a SAAS application that enables registered customers to deploy data platform on their AWS environment using pulumi. To do that, our SaaS application needs to programmatically create a pulumi account when a customer is registered at our application. Question: Are there pulumi APIs we can call to do the followings? • Organization management APIs (Create, update, etc.) • User / Profile management APIs (Create, update, associate a user to Organization, etc.) • Access tokens APIs (This can be part of User APIs)

Is it possible to suppress the

[DEBUG]

diagnostics? e.g.

Diagnostics:
  pulumi:pulumi:Stack (...):
    2020/02/24 11:50:18 [DEBUG] Registering Data Sources for "Analysis Services"..
    2020/02/24 11:50:18 [DEBUG] Registering Resources for "Analysis Services"..
    2020/02/24 11:50:18 [DEBUG] Registering Data Sources for "API Management"..
    2020/02/24 11:50:18 [DEBUG] Registering Resources for "API Management"..
    2020/02/24 11:50:18

My google-fu is clearly failing me

Hello, Is there any way to print just the errors from the stack after the stack is completed(like passing additional command arg or rest api) or do I need to dump into file and apply unix command to do it?

Courtesy of @prehistoric-application-36248

Its not finished, but figured I would show it off a bit because its coming along! 💪

👋 Took some artistic liberties with the theme to make it pop a bit more. Just working on getting the editor itself to match as well as testing as many languages as I can. Feedback is appreciated!

Hi, I'm having an issue using Pulumi as part of a Codefresh continuous integration pipeline. We can get it working, but then every so often pulumi upgrades and our builds all begin failing with messages like

pulumi:providers:aws default_1_22_0  error: no resource plugin 'aws-v1.22.0' found in the workspace or on your $PATH, install the plugin using pulumi plugin

Is there a way to pin pulumi and the plugins to particular versions so that we can choose when to do the update and not be forced into it? The step config we're using currently is:

pulumi_deploy_production:
    description: Deploy to production
    type: freestyle
    stage: release
    image: pulumi/pulumi:v1.7.1
    working_directory: "infrastructure/pulumi/production"
    commands:
      - cd ../common && npm install
      - cd ../production && npm install
      - pulumi plugin install resource aws v1.15.0
      - pulumi stack select MachineLabsLtd/production --non-interactive
      - pulumi up --non-interactive --skip-preview --yes

Hi all, Is there a way to use/add custom CRDs in Pulumi-query-Kubernetes? https://www.pulumi.com/blog/query-kubernetes/

I'm new to Pulumi. I've noticed that pulumi.all is equivalent to Promise.all and .apply() is similar to .then(). What seems to be missing is an equivalent for await. That feels like a step back in time to using chained .then() statements. I've noticed I can wrap it in a promise sort of like this:

return new Promise(resolve => return pulumi.output(variable_name).apply(val => {resolve (val);})

The returned object looks correct, though Typescript seems to have lost all concept of the object types being returned. This makes me think about the AWS SDK where you can add .promise() to the end of a number of calls to work cleanly with await. I'm far from being a Typescript guru, but how difficult is it to have something similar in Pulumi?

Is there a way to listen to pulumi completing the change for a resource? I want to be able to update GitHub using https://developer.github.com/v3/repos/deployments/#create-a-deployment-status so that when I deploy a new image for a Deployment, I can update GitHub on this. But I need to be able to know that the changes were completed for a specific resource.

When installing a Helm chart with Pulumi, is it possible to mark some of the Kubernetes resources it creates as “ignore-changes”?

Hello!!! I’m new to Pulumi. Is there a testing facility for Dynamic providers?

Hi! I am having some issue deploying EKS. I initially tried deploying to us-east-1. It was able to deploy some resources but not the cluster (the cluster could not be provision for lack of capacity). So I changed the region and tried again. Doing this it tried to change some resources to the new region. It was able to successfully create the cluster resources and the cluster is up and running but it completed with error. Now the state has 2 resources (eksClusterInternetEgressRule and eksClusterSecurityGroup) that it tries to delete everytime I run a

pulumi up

but it fails everytime with a

No security group with ID "XXX"

error message. I am not able to modify my infrastructure because of those 2 resources. I tried removing the resources from the state by specifying the urn and it can’t find them either…

Hi. I saw your Iac workshop on GitHub (https://github.com/pulumi/infrastructure-as-code-workshop). Is it okay for you if I reuse the content (slides and lab) for an internal presentation / workshop in my company ?

Hello, Today, I’m able to publish pulumi component in a private npm repository in order to reuse them in several pulumi projects, and it works like a charm (thanks pulumi team). As the component are in typescript, we can only reuse them in pulumi typescript project. In a big organisation, we would be able to let some team works in other languages like python. Is there a way to share component for a multi-language use ? May be by generating them ?

I am trying to re-parent a resource into a custom component. The old parent is the stack (which is the implicit root). On my resource, which is now inside the

pulumi.ComponentResource

subclass, I add these custom resource options:

parent: this,
aliases: [{ parent: pulumi.getStack() }]

However, I still see a

create

and

delete

as separate lines in my Pulumi preview. How can I make this work?

Hi, What is the expected behavior of an

pulumi.config.require_object("config")

if the

config

object has data that was deemed a secret. I am trying to understand if

require_object

automatically translates the portion of the object into a secret that was set by using pulumi config set --secret. FWIW it seems to mask the secret in pulumi preview, but when writing the state file it seems to be showing keys in plain text.

@early-intern-90238 please check your inbox

👋 hi all

I’m banging my head against the wall.. I’m trying to get value interpolation working but nothing seems to work… I read the https://www.pulumi.com/docs/intro/concepts/programming-model/ a bunch of times… and I still don’t understand why I can’t get values showing up. I tried using

.all().apply()

,

.output()

,

.interpolate()

, … idk what to do..

@ripe-lighter-13633 try putting the "push" inside the apply() clause. Make your mToPush just a variable and ignore the return of .all(). Either that, or you may need to "return" a value from the apply().

I had that previously and it still didn’t work

I edited my message. Not sure if you saw the update?

after putting .push() inside apply, I get an empty results

probably because it’s all happening async

Can you paste in that revised snippet?