I have a k8s cluster, and I saved another k8s cluster’s kubeconfig in the first k8s cluster as secret. I want to build a k8s provider from the kubeconfig secret.

export const secret = k8s.core.v1.Secret.get("kubeconfig", "admin/kubeconfig-for-admin");
export const k8sProvider = new k8s.Provider("new-cluster", {
    kubeconfig: secret.stringData["kubeconfig"],
});

But the

k8sProvider

does not work as expected, I couldn’t see the kubeconfig field in it. Does anyone know why it is that? I’m pretty sure the kubeconfig was stored as base64 string in the

admin/kubeconfig-for-admin

.

Upgrade Free Plan to Team Starter - Is it possible to do this without creating a new organization?

Hi community, a new one here! 🙂 I'm a full stack-dev working for a social enterprise named Kinder.world - we have just started to use Pulumi recently

So here is my first question: feel free to direct me to the corresponding docs or FAQ item if I missed something 🙂 What is your best practice to manage / organise stacks and the code? Specifically: I created a script which manage my dns zones in AWS Route53. I also saved this to a git repo, done. Now my next thing would be to create another script to manage docker containers in ECS. But where should I put this? I am afraid I will lock myself if I create everything in the same stack / repo, because when I intend to delete some portion of my resources (let’s say ECS this case) with `pulumi destroy`I would kill everything (including my dns zones). Do you usually create new repo / stack for each resources?

If I use the local state backend with

pulumi login file://<path>

, does

<path>

represent an absolute path, or can I use a relative one too? If so, how? I tried

file:./state

, but I only get

.pulumi

folder created in my current folder, not in a

state

subfolder.

pulumi up

is getting stuck at the preview stage, even if the pulumi program is empty:

$ pulumi up
Previewing update (dev):
     Type                 Name               Plan     
     pulumi:pulumi:Stack  empirica-base-dev           
 
System Messages
  ^C received; cancelling. If you would like to terminate immediately, press ^C again.
  ^C received; terminating
  ^C received; terminating

It won't even terminate with Ctrl+C. How do I troubleshoot this? I did try reinstalling pulumi. Also tried creating a new stack. Still fails.

We are trying to deploy multiple pulumi stacks in parallel while there are stack references defined among those stacks. We defined internal guidelines to make sure we never make pulumi changes that change the value of the pulumi resources that are being referenced by another stack, which means during the paralel build of a set of pulumi stacks, it doesn’t matter if one stack is referencing an older state of another stack or a newer state. But we are wondering if pulumi actually support reading the stack reference while that stack is being deployed by another pulumi process?

should pulumi work on node v 13.2?

i can’t even get the basic “getting started” with AWS to deploy

Hi all! Pulumi n00b here, but I'm moving right along. I basically have an entire vSphere datacenter automated with Python and Pulumi. The only problem that I'm seeing is vSphere HA is trying to configure on the hypervisors before datastores have been created and mounted and thus failing. I'm not seeing a

dependsOn

flag in

opts=

for the vSphere cluster resource. Looking through the docs that's where I see the flag but I don't see it in vSphere.

vSphere Cluster HA error.

Hello All. Is there any way we can provision the node while it's creating by Pulumi ? I am referring Terraform provisioner here.

• I get the same errors during the initial deployment as well as updates. • Other resources are consistently deploying without issues • Pulumi fails on the exception, then AWS successfully deploys the Lambda, leaving Pulumi without the correct state • Deployments using the same stack, branch, and AWS credentials from other machines/connections do not face the same error • I started having this issue with an older 1.x version and have since updated to 2.1.0. • There was a thread started a few days ago that looks like the same issue. https://pulumi-community.slack.com/archives/CJ909TL6P/p1588234271145800

deleteBeforeReplace

doesn't solve the issue. Here is the index.js code that I'm producing the error with

const aws = require( '@pulumi/aws' );
const awsx = require( '@pulumi/awsx' );

const api = new awsx.apigateway.API( 'some-name', {
	routes: [ {
		path: '/something',
		method: 'POST',
		eventHandler: new aws.lambda.CallbackFunction( 'lambda-test', {
			memorySize: 128,
			callback: ( event ) => {
				console.log( event );
			},
			description: 'Some description',
		} ),
	} ],
} );

return {
	url: api.url,
};

Hi all. As of last Friday, I've started getting an error anytime I try to create or update a Lambda function.

Error creating Lambda function: ResourceConflictException: Function already exist

Here are relevant log entries from a Pulumi update that shows a ResourceConflictException happens 5 seconds before the successful deployment, possibly suggesting an issue with timeouts and retries.

@timestamp,@message
2020-05-05 18: 07: 15.676, "{""eventVersion"":""1.05"",""userIdentity"":{""type"":""IAMUser"",""principalId"":""AIDA3SNA767QOR3RN3I4Y"",""arn"":""arn:aws:iam::795443460064:user/chad.ort.snopes.local"",""accountId"":""795443460064"",""accessKeyId"":""AKIA3SNA767QJLIPDJSB"",""userName"":""chad.ort.snopes.local""},""eventTime"":""2020-05-05T17:50:10Z"",""eventSource"":""<http://lambda.amazonaws.com|lambda.amazonaws.com>"",""eventName"":""CreateFunction20150331"",""awsRegion"":""us-west-2"",""sourceIPAddress"":""97.114.75.236"",""userAgent"":""aws-sdk-go/1.29.7 (go1.13.8; darwin; amd64) APN/1.0 HashiCorp/1.0 Terraform/0.11+compatible (+<https://www.terraform.io>)"",""requestParameters"":{""functionName"":""lambda-test-2d7ed19"",""runtime"":""nodejs12.x"",""role"":""arn:aws:iam::795443460064:role/lambda-test-f532b55"",""handler"":""__index.handler"",""code"":{},""description"":""write events to queue and return 200"",""timeout"":180,""memorySize"":128,""publish"":false,""environment"":{}},""responseElements"":{""functionName"":""lambda-test-2d7ed19"",""functionArn"":""arn:aws:lambda:us-west-2:795443460064:function:lambda-test-2d7ed19"",""runtime"":""nodejs12.x"",""role"":""arn:aws:iam::795443460064:role/lambda-test-f532b55"",""handler"":""__index.handler"",""codeSize"":7957025,""description"":""write events to queue and return 200"",""timeout"":180,""memorySize"":128,""lastModified"":""2020-05-05T17:50:10.334+0000"",""codeSha256"":""Hcc9msmvBw+tYRsWLEEf1wItAJ6UEbRtCXg6QwD49ds="",""version"":""$LATEST"",""environment"":{},""tracingConfig"":{""mode"":""PassThrough""},""revisionId"":""275fde3f-46de-4790-bd08-90f2be49e309"",""state"":""Active"",""lastUpdateStatus"":""Successful""},""requestID"":""37d15659-0d93-4586-80ba-dfce76c82a94"",""eventID"":""47ff845a-105f-4d51-a06b-5c82bed05e57"",""eventType"":""AwsApiCall"",""recipientAccountId"":""795443460064""}"
2020-05-05 18: 07:10.282, "{""eventVersion"":""1.05"",""userIdentity"":{""type"":""IAMUser"",""principalId"":""AIDA3SNA767QOR3RN3I4Y"",""arn"":""arn:aws:iam::795443460064:user/chad.ort.snopes.local"",""accountId"":""795443460064"",""accessKeyId"":""AKIA3SNA767QJLIPDJSB"",""userName"":""chad.ort.snopes.local""},""eventTime"":""2020-05-05T17:51:04Z"",""eventSource"":""<http://lambda.amazonaws.com|lambda.amazonaws.com>"",""eventName"":""CreateFunction20150331"",""awsRegion"":""us-west-2"",""sourceIPAddress"":""97.114.75.236"",""userAgent"":""aws-sdk-go/1.29.7 (go1.13.8; darwin; amd64) APN/1.0 HashiCorp/1.0 Terraform/0.11+compatible (+<https://www.terraform.io>)"",""errorCode"":""ResourceConflictException"",""errorMessage"":""Function already exist: lambda-test-2d7ed19"",""requestParameters"":{""functionName"":""lambda-test-2d7ed19"",""runtime"":""nodejs12.x"",""role"":""arn:aws:iam::795443460064:role/lambda-test-f532b55"",""handler"":""__index.handler"",""code"":{},""description"":""write events to queue and return 200"",""timeout"":180,""memorySize"":128,""publish"":false,""environment"":{}},""responseElements"":null,""requestID"":""6f0ea757-b48e-4f43-8de2-0d92be3bd2c3"",""eventID"":""a3e9e0ac-643a-41e4-acb4-e26330d9c213"",""eventType"":""AwsApiCall"",""recipientAccountId"":""795443460064""}"

Additional known info

Is there any way to force all child or dependent resources to be replaced when a parent resource is replaced?

Hello! Can someone show me how to get the output of a docker container? I’ve tried for example this, but it just gives me undefined:

myContainer.containerLogs.apply(logs => console.log("container logs:", logs));

Hello, do we know when

@pulumi/kubernetesx

will be upgrading to the last pulumi version? Still depending on pulumi 1.12

This seems like a silly question, but how do you invoke Pulumi actions (up/preview/destroy) through the Go SDK? I see the implementation in, for example,

./pkg/cmd/pulumi/up.go

but I can't seem to find an example of not using the CLI tool to update/preview -- I was hoping to do it through a non-interactive service.

Hi! I just tried to create a service account and assign a role to it in GCP and it worked. However, running

pulumi destroy

didn’t only delete the service account in question, but also blew away the role assignments from ALL service accounts that had identical role. Am I doing this wrong? Seems quite risky. Here’s the code I used to create the service account and assign the role:

const serviceAccount = new gcp.serviceAccount.Account("myServiceAccount", 
    {
     accountId: SID,
     displayName: SNAME,
     description: "Service account Created by Pulumi"
    }
);

const iam = new gcp.projects.IAMBinding("myBinding",
   {
        members: [pulumi.interpolate `serviceAccount:${serviceAccount.email}`],
        role: "roles/storage.objectViewer",  
   }
);

I can deploy, but I'm still facing this error message while deploying. I've installed previously python3-dev, python-dev and_build_essentials packages. I've tried also to add wheel package in requirements.txt. I'm working in linux mint 19.03 (ubuntu 18.04). Any Idea?

Building wheels for collected packages: pulumi-aws, grpcio
  Running setup.py bdist_wheel for pulumi-aws ... error
  Complete output from command /home/ayeste/Documentos/Pulumi/aws-python/venv/bin/python3 -u -c "import setuptools, tokenize;__file__='/tmp/pip-build-zkdm1y8t/pulumi-aws/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" bdist_wheel -d /tmp/tmpqfs9agflpip-wheel- --python-tag cp36:
  /usr/lib/python3.6/distutils/dist.py:261: UserWarning: Unknown distribution option: 'long_description_content_type'
    warnings.warn(msg)
  usage: -c [global_opts] cmd1 [cmd1_opts] [cmd2 [cmd2_opts] ...]
     or: -c --help [cmd1 cmd2 ...]
     or: -c --help-commands
     or: -c cmd --help
  
  error: invalid command 'bdist_wheel'
  
  ----------------------------------------
  Failed building wheel for pulumi-aws
  Running setup.py clean for pulumi-aws
  Running setup.py bdist_wheel for grpcio ... error
  Complete output from command /home/ayeste/Documentos/Pulumi/aws-python/venv/bin/python3 -u -c "import setuptools, tokenize;__file__='/tmp/pip-build-zkdm1y8t/grpcio/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" bdist_wheel -d /tmp/tmpgydsg87cpip-wheel- --python-tag cp36:
  Found cython-generated files...
  usage: -c [global_opts] cmd1 [cmd1_opts] [cmd2 [cmd2_opts] ...]
     or: -c --help [cmd1 cmd2 ...]
     or: -c --help-commands
     or: -c cmd --help
  
  error: invalid command 'bdist_wheel'
  
  ----------------------------------------
  Failed building wheel for grpcio
  Running setup.py clean for grpcio
Failed to build pulumi-aws grpcio
Installing collected packages: protobuf, grpcio, pulumi, arpeggio, attrs, parver, semver, pulumi-aws
  Running setup.py install for grpcio ... \
done
  Running setup.py install for pulumi-aws ... done
Successfully installed arpeggio-1.9.2 attrs-19.3.0 grpcio-1.28.1 parver-0.3.0 protobuf-3.11.3 pulumi-2.1.0 pulumi-aws-2.2.0 semver-2.9.1

it might be a naive question, but is there a way to launch pulumi as an interactive console?

hi there. in AWS: for Eks nodegroup pulumi creates “managed nodegroup” or “self-managed nodegroup”? I’m looking for a way to create self-managed nodegroup.

Hi, I've got an odd question and maybe there's a much better way to do this. Essentially I'm trying to implement a different comparison for my ECS service's task definition in golang. Our Docker build process produces images with different IDs even when the base image and the service we're running in the image have not changed. This causes Pulumi to always create a new task definition during deployments. We want to avoid this so I was looking into computing the hash of the image inputs (base image and service executable) and comparing that to the hashes of the currently deployed image. If they are the same, I want the task definition to use the same Docker image ID instead of the new one. To accomplish this I need to store those hashes and, ideally, I'd like to store them in the state file. I've tried adding them as outputs on my component resource that encapsulates all the ECS resources but I can't find a way to read those values at runtime of the pulumi program. I tried using

pulumi.Context.ReadResource

but this requires my component resource to embed

CustomResourceState

instead of

ResourceState

and when calling

ReadResource

it produces an error stating there isn't a plugin for the namespace I'm using for my component resource. So, finally, to my question, is there a way to read the outputs of a component resource inside a pulumi program without just parsing the state file directly?

Hello, i have a question I'm trying to run pulumi on Gitlab CI with Azure Blob backend and Azure KeyVault. I'm authenticating using service principal. For some reason when i try to select a stack, pulumi just hangs, and doesn't produce any output or error, even with

-v=9

Question on the languages support by Pulumi. Is anyone has any opinion on the support between TypeScript and Python. I’m wonder if there is still something which is working in Typescript which is not available in Python.

What is the easiest way to print instantly during pulumi up (using nodejs)?

Q: Is there a way to make Pulumi insert auto-generated hex string in the middle of a resource name? So that, for example, instead of

package-385ac3

I could have

package-385ac3.zip

(in case with GCP

BucketObject

resource)

Hi, my team currently supports several teams with Terraform modules and we plan to switch to Pulumi instead. We have already completed most of the stuff for the new setup: wrote a compoment resource, npm packaging, CI/CD, state backend, etc. The only thing stopping us is the migration of existing resources. I had a look at the import function and it seems pretty cool, but it got me thinking about how to do that for a component resource. The first thing I thought of was adding parameters to the component resource to pass in the IDs of the resources. I'm not sure if that's the best approach to the problem. Do you guys have any experience or solutions for the migration?

Still trying to figure out what

pulumi up

freezes everytime. Can someone familiar with go tell me what

GetRequiredPlugins

is doing? It seem to be the function that is stuck.

$ pulumi up --logtostderr -v=9
I0507 13:47:42.755386   28275 backend.go:410] found username for access token
I0507 13:47:43.697486   28275 backend.go:410] found username for access token
Previewing update (dev):
I0507 13:47:44.093878   28275 backend.go:410] found username for access token
I0507 13:47:44.093702   28275 backend.go:887] Stack dev being updated to version 1
I0507 13:47:44.476166   28275 plugins.go:76] gatherPluginsFromProgram(): gathering plugins from language host
I0507 13:47:44.476447   28275 plugins.go:426] GetPluginPath(language, nodejs, <nil>): found on $PATH /home/gsuess/.pulumi/bin/pulumi-language-nodejs
I0507 13:47:44.476491   28275 plugin.go:83] Launching plugin 'nodejs' from '/home/gsuess/.pulumi/bin/pulumi-language-nodejs' with args: 127.0.0.1:39367
I0507 13:47:44.553950   28275 langruntime_plugin.go:170] langhost[nodejs].GetPluginInfo() executing
I0507 13:47:44.554275   28275 langruntime_plugin.go:83] langhost[nodejs].GetRequiredPlugins(proj=leaderboard,pwd=/home/gsuess/meteor-leaderboard,program=.) executing
     Type                 Name             Plan     
     pulumi:pulumi:Stack  leaderboard-dev           
 
System Messages
  ^C received; cancelling. If you would like to terminate immediately, press ^C again.
  ^C received; terminating

I've tried to isolate the problem, but could not bisect it (seems to be very prone to some cache somewhere). Related thread about past attempts to get this problem isolated here: https://pulumi-community.slack.com/archives/C84L4E3N1/p1588696231371300 Any help would be really appreciated. Can't figure this one out so far.

Hello there... I'm trying to provide some recommendations internally (and in a blog post) about Pulumi SKUs, pricing and limits. What I'm stuck on is the Community SKU does not mention project stack limits, Team Starter SKU mentions 20 project stacks, and Team Pro SKU removes any limits. 1. What does project stack limit actually mean? Projects or stacks, or stack count across all projects. 2. What is the limit for community?