pulumi-community #general

Docs

Join the conversation Join Slack

general

loud-scientist-13640

11/30/2018, 4:36 PM

Any word on when 0.16.6 for Windows will be available to download, current error message

cold-coat-35200

11/30/2018, 5:19 PM

Hi, I'm trying to restructure my pulumi repo to contain 2 pulumi project with shared custom modules and shared node_modules:

├── bastion // pulumi project
├── dliver //pulumi project
├── modules // shared custom modules
├── node_modules // shared node_modules
└── static

bastion/dliver has it's own package.json and pulumi related files, but their node_modules directory is empty, they only using the shared node_modules. It work fine, until I'm trying to use a custom dynamic provider, in that case I got this error during the preview:

Diagnostics:
  pulumi:providers:pulumi-nodejs (default):
    error: could not read plugin [/home/ncsibra/.pulumi/bin/pulumi-resource-pulumi-nodejs] stdout: EOF
 
  pulumi:pulumi:Stack (bastion-bastion-dev):
    (node:2514) ExperimentalWarning: queueMicrotask() is experimental.
 
    internal/modules/cjs/loader.js:589
        throw err;
        ^
    Error: Cannot find module '/home/ncsibra/dev/prmrgt-infra/pulumi/bastion/node_modules/@pulumi/pulumi/cmd/dynamic-provider'
        at Function.Module._resolveFilename (internal/modules/cjs/loader.js:587:15)
        at Function.Module._load (internal/modules/cjs/loader.js:513:25)
        at Function.Module.runMain (internal/modules/cjs/loader.js:760:12)
        at startup (internal/bootstrap/node.js:303:19)
        at bootstrapNodeJSCore (internal/bootstrap/node.js:872:3)

The module is in the

/home/ncsibra/dev/prmrgt-infra/pulumi/node_modules/@pulumi/pulumi/cmd/dynamic-provider

directory, not in the

/home/ncsibra/dev/prmrgt-infra/pulumi/bastion/node_modules/@pulumi/pulumi/cmd/dynamic-provider

. Without the dynamic provider, the preview runs fine. I think the problem is not the dynamic provider, because with the old structure(it was one pulumi project, which contained the modules and the node_modules directory too with 1 package.json) everything worked fine. Any idea why it's not find the

@pulumi/pulumi/cmd/dynamic-provider

module, but find everything else in

@pulumi/pulumi

and

@pulumi/aws

bitter-oil-46081

11/30/2018, 6:18 PM

I believe I understand what is going on here. The dynamic provider’s launch script uses a relative path (from the CWD) into ./node_modules instead of doing something like a require, which would allow node to find it. This should be easy enough to fix both in the product and for you to unblock yourself. Give me a moment to try some stuff.

bulky-policeman-29913

11/30/2018, 6:46 PM

okay what am I missing you have to use a pulumi account? I am trying my first new project and it wants a login but I thought you could just manage things locally/?

bulky-policeman-29913

11/30/2018, 6:49 PM

guess --offline I am confused why it wants github permissions

orange-tailor-85423

11/30/2018, 7:25 PM

just a general GCP tip - careful with IAM bindings vs IAM member. Unintended consequences 🙂

👍 1

tall-monitor-77779

11/30/2018, 7:26 PM

further to this, I get

Cannot find module 'spdx-license-ids' from 'scan.js'

when running tests via jest https://pulumi-community.slack.com/archives/C84L4E3N1/p1543587290005700

most-pager-38056

11/30/2018, 8:52 PM

Hi! I’m trying to use Github Actions to deploy stack changes automatically, but the action is failing without any message:

### STARTED Pulumi Deploy (Current Stack) 20:40:29Z

(pulling, digest, status, etc…)

### FAILED Pulumi Deploy (Current Stack) 20:40:30Z (97ms)

I can run

pulumi update/preview

locally and i’ve setup my

.pulumi/ci.json

and the workflow secrets correctly, so i’m not sure how to debug this. ☹️

billowy-television-23506

12/01/2018, 10:12 PM

Does pulumi have a transpiler for ARM templates?

faint-motherboard-95438

12/03/2018, 3:30 PM

Hello here, It seems I got something here. I use

gcp.container.Cluster

to create and manage a cluster with pulumi. I’m still in the learning phase, so I’m doing a lot of trials and errors so I have to update and rebuild the stack often. I changed the way the cluster is named and as expected, pulumi decided to update it (actually, replaced it). But instead of having my new cluster up and running I have now 2 clusters, the new and the old one. I suppose the old one should have been deleted in a replace operation, but it’s actually not (and no error reported about any issue that would have prevented it). I’m well aware that’s not really a daily use case, but anyway, it feels like a bug.

bulky-policeman-29913

12/03/2018, 7:57 PM

Can someone explain what I am missing here

bulky-policeman-29913

12/03/2018, 7:58 PM

/home/user/dev/pulumi/bean/node_modules/@pulumi/pulumi/config.js:141
source-map-support.js:419
            throw new ConfigMissingError(this.fullKey(key));
                  ^
Error: Missing required configuration variable 'aws:region'
source-map-support.js:422
	please set a value using the command `pulumi config set aws:region <value>`
    at Config.require (/home/svause/dev/pulumi/bean/node_modules/@pulumi/pulumi/config.js:141:19)
    at exports.region.utilities.requireWithDefault (/home/svause/dev/pulumi/bean/node_modules/@pulumi/config/vars.ts:44:78)
    at Object.requireWithDefault (/home/svause/dev/pulumi/bean/node_modules/@pulumi/utilities.ts:45:16)
    at Object.<anonymous> (/home/svause/dev/pulumi/bean/node_modules/@pulumi/config/vars.ts:44:44)
    at Module._compile (internal/modules/cjs/loader.js:688:30)
    at Object.Module._extensions..js (internal/modules/cjs/loader.js:699:10)
    at Module.load (internal/modules/cjs/loader.js:598:32)
    at tryModuleLoad (internal/modules/cjs/loader.js:537:12)
    at Function.Module._load (internal/modules/cjs/loader.js:529:3)

bulky-policeman-29913

12/03/2018, 7:58 PM

I have my yaml files but in debug it would appear they cannot be read?

elegant-crayon-4967

12/03/2018, 9:48 PM

New to Pulumi...following the PulumiTV YouTube Episode 001. What's crazy is the update successfully said it created an S3 bucket, but it shouldn't have been able to do that. I haven't authenticated with MFA or setup my credentials file. I wonder where the S3 bucket actually is that I so called "created"?🤔

adventurous-stone-37349

12/03/2018, 10:03 PM

Dropping in here - I had a chance to talk to @incalculable-sundown-82514 at DockerCon EU and am looking forward to helping work on a provider for Packet (http://packet.com). We have existing Terraform support already.

👋 3

❤️ 3

adamant-restaurant-73893

12/03/2018, 10:19 PM

You wanted to be able to use GitLab to login to Pulumi? Now you can https://blog.pulumi.com/welcoming-gitlab-users-to-pulumi - @clever-sunset-76585 has the details

👏 2

👍 3

most-pager-38056

12/03/2018, 10:45 PM

Hello! 👋 I’m still trying to automate our infrastructure using Github Actions. 😁 I’m trying to use Pulumi to create and push an image to ECR, but i can’t authenticate to ECR from a Github Action. Locally, i can inspect and push the image normally, so i’m not sure if there are any differences between how Pulumi handles the registry authentication locally vs using a CI provider.

faint-motherboard-95438

12/03/2018, 11:09 PM

Guys, I could use some help, I have kind of an egg and chicken problem here. Right now I’m creating and managing the cluster inside the pulumi stack, then somewhere later in the deployment I’m creating roles and bindings for some services, but that can’t work unless I’ve created a RoleBinding that grants my current google identity the role

cluster-admin

for this cluster beforehand (as stated here : https://cloud.google.com/kubernetes-engine/docs/how-to/role-based-access-control#setting_up_role-based_access_control), see where I’m going with that ? I need the cluster to be created to be able to make this binding before being able to create any other Role or ClusterRole, but since the cluster creation and the roles are part of the automation, the only way everything works as expected (in one shot) would be to create this RoleBinding right after the cluster is ready inside the pulumi stack deployment. Doing so would require to get my current active google identity to make the right binding on the right account. How/Can I do that within pulumi with the

@pulumi/kubernetes

package (I’m using the typescript flavor) ?

billowy-television-23506

12/03/2018, 11:45 PM

I have an ARM template and in the resource of the type

Microsoft.Web/sites

in the

siteConfig

section there is:

"alwaysOn": true,
"linuxFxVersion": "DOCKER CONTAINER"

The Pulumi SDK supports

alwaysOn

is there a way to pass “linuxFxVersion”?

damp-pillow-67781

12/04/2018, 1:08 AM

Hey anyone know why this happens?

Plan apply failed: unable to fetch resource description for extensions/v1beta1: Unauthorized

prehistoric-analyst-52684

12/04/2018, 11:15 AM

I have a bunch of terraform configured resources. I want to convert everything to Pulumi. Creating the code seems to be the easier part, but how do I import the resources in the terraform state file into checkpoint data? I know that Pulumi does not yet have an import command like terraform. What do people do to import stuff into the checkpoint?

faint-motherboard-95438

12/04/2018, 11:20 AM

this error is not really helpful :

error: Error: Invoke of 'gcp:serviceAccount/getAccount:getAccount' failed: undefined (undefined)
        at Object.<anonymous> ([...]/node_modules/@pulumi/pulumi/runtime/invoke.js:82:23)
        at Generator.next (<anonymous>)
        at fulfilled ([...]/node_modules/@pulumi/pulumi/runtime/invoke.js:17:58)

coming from :

const account = gcp.serviceAccount.getAccount({ accountId: '[account_id]' })

I followed the function declaration from

getAccount.d.ts

, I’m missing something here

loud-scientist-13640

12/04/2018, 3:26 PM

Would like to know when the GitHub Pulumi App will have support for Azure Pipelines? Currently it states: "Currently, Pulumi’s GitHub application only supports Travis CI. (For instructions on configuring Travis, see our guide.) If you would like GitHub workflow integration for another CI system, please let us know."

full-dress-10026

12/04/2018, 10:51 PM

I have this code which is supposed to create a SNS subscription:

let alertsTopicSubscription = new aws.sns.TopicSubscription("alerts-lambda-subscription", {
    topic: alertsSnsTopic,
    protocol: "lambda",
    endpoint: "my-lambda-arn" 
});

However, if I go to the AWS console, the Lambda function does not have the SNS topic listed as a trigger. Is there another step I need to perform to get this to work as expected?

full-dress-10026

12/05/2018, 12:31 AM

Turns out I needed to create a

aws.lambda.Permission

first. Is there a nice API to do both of these things?

faint-motherboard-95438

12/05/2018, 6:51 PM

How can one access properties of a provider ? In my case I’d like to access the

credentials

property of the

gcp

package

Provider

I pass down my components.

adamant-restaurant-73893

12/05/2018, 7:04 PM

Blogging and a downloadable white paper on our views on delivering Cloud Native Infrastructure as Code https://blog.pulumi.com/delivering-cloud-native-infrastructure-as-code-a-pulumi-white-paper

👍 3

cuddly-barista-79466

12/05/2018, 8:18 PM

Forgive my lack of understanding the Pulumi Terraform connection, but if I wanted to use the Terraform provider in Pulumi how would that work if there is no pulumi provider/resource for BigIP yet? https://github.com/terraform-providers/terraform-provider-bigip

cuddly-barista-79466

12/06/2018, 2:23 PM

I think I was oversold on Pulumi capabilities vs Terraform for On-Prem IAC. Finding it very difficult to understand the implementation of Terraform stuff in Pulumi to get the like for like selling point of Pulumi. Can someone honestly tell me if I can use Pulumi to: - Use VSPhere for VM Create - Use Chef to configure VM - Use F5 for Loadbalancing - Use Infoblox for DNS These are all supported Providers and or Provisioners. Vsphere being the only package available for VM Creation in Pulumi. Not even sure how Terraform provisioners are referenced in Pulumi or if they even will be…. Is there a time frame for such implementations?

full-dress-10026

12/06/2018, 9:34 PM

An AWS lambda function created externally and referenced in my Pulumi code via

Function.get

was deleted externally. I have commented out the

Function.get

code but now Pulumi is stuck when running

pulumi refresh

with:

Preview failed: refreshing urn:pulumi:app-prod::app::aws:lambda/function:Function::send-alert-lambda: InvalidParameter: 1 validation error(s) found.
    - minimum field size of 1, GetFunctionInput.FunctionName.

Title

full-dress-10026

12/06/2018, 9:34 PM

An AWS lambda function created externally and referenced in my Pulumi code via

Function.get

was deleted externally. I have commented out the

Function.get

code but now Pulumi is stuck when running

pulumi refresh

with:

Preview failed: refreshing urn:pulumi:app-prod::app::aws:lambda/function:Function::send-alert-lambda: InvalidParameter: 1 validation error(s) found.
    - minimum field size of 1, GetFunctionInput.FunctionName.

white-balloon-205

12/06/2018, 11:54 PM

Cc @microscopic-florist-22719 @stocky-spoon-28903.

View count: 1