Hey guys, I've got a question regarding Azure subscriptions. What's the best method for migrating existing stacks from one sub to another? At the time, I'm thinking

destroy

+

update

. On the other hand, Azure supports resources migration (through portal) but I'm guessing I'll need to edit the resources IDs in the stack files with the new subscription ID, is it worth the try? Thanks

This may sound weird but whos playlist is that playing the music on Pulumis site? I want that for my heads down coding sessions lol. Its so good.

Hi folks! Just started setting up my first pulumi project. Where would be a good spot to ask for help around integration testing in JS?

Hi guys! A quick question, maybe someone already investigated it. Suppose we have two pulumi stacks, A and B, self-hosted backend. B uses StackReference to get some outputs from A. With new (awesome!) PULUMI_SELF_MANAGED_STATE_LOCKING=1 we can be sure that if we try run concurrently updates for A or B, we won't get into trouble because the state of particular stack would be locked. However the question is, would we be protected from this kind of trouble if we run updates at the same time for A and B (which uses values from A)? Would really appreciate some insight on the matter, because I didn't find any info in docs 🙂

General question, imagine i have a stateful set with 500gb of data on a PVC/PV. And i need to duplicate that data to a bunch of other statefulsets. Apart from zipping and putting the data on s3 and loading it on deploy. is there a ’better way”, something like snapshots or something?

If I'm using s3 as a backend, is there a way of not doing a pulumi login s3:// and just specify it in a env var or pulumi argument ex:

PULUMI_BACKEND=<s3://mybucket> pulumi up -s mystack

Hi everyone! Is it possible to prevent one stack from getting destroyed as long as another stack is up? E.g. I have stack A which creates an EKS cluster and stack B for creating k8s resources in the cluster. Destroying stack A when stack B is still up leaves that stack in a bad state. And more worse, there may be unattended resources left that are not under Pulumi's control at all, e.g. load balancers that have been created by the load balancer manager. Can stack B register itself in stack A so that stack A can't be destroyed until stack B deregisters itself on its own destruction?

Hi, I would like to extract all the resource attributes when running

unittest

. I have prepared the following 2 files: https://pastebin.com/qnykN38V and https://pastebin.com/vcWbyn3Y. On running

python3 -m unittest test.py

I get:

{<pulumi.output.Output object at 0x7ff78348bb80>: <pulumi.output.Output object at 0x7ff78343f670>, <pulumi.output.Output object at 0x7ff783497d00>: <pulumi.output.Output object at 0x7ff783458490>, <pulumi.output.Output object at 0x7ff7834a5730>: <pulumi.output.Output object at 0x7ff78346df40>, <pulumi.output.Output object at 0x7ff7834a51f0>: <pulumi.output.Output object at 0x7ff783404a00>, <pulumi.output.Output object at 0x7ff7834b0490>: <pulumi.output.Output object at 0x7ff783415af0>}

. How do I force pulumi to return the Output? Thanks

I'm trying to revive an old project that had examples tests written with the first go integration test sdk. How do I get around the

Failed to load gRPC binary module because it was not installed for the current system

try running npm rebuild

errors?

Hi sorry if this is a FAQ, I’ve done some searching and haven’t found anything. As a Terraform user who’s new to Pulumi, I’m surprised to see that the graph of my stack has only two levels. There’s just the stack, and then it depends on every other resource. And in the graph I don’t see any relationships between any of the resources. This is surprising to me because Terraform automatically registers the relationships (dependencies) between resources and then renders a more useful graph. Am I missing something?

As someone new to Pulumi who create a TS project and is now working on a Python project, I’m wondering why the Python SDK doesn’t seem to include

interpolate

and whether it might get it (or something equivalent) at some point?

Hi again, I am trying to follow the advice of splitting a 'monolitic' infra into 'mini-projects'... So I've started to move out the lambdas into a 'funtions' projects and then leave everything else into an 'infrastructure' project. If I pulumi up in infra I delete the functions, that's good, and expected. When I pulumi up in the functions project I reference the required infrastructure, all good, create the functions, good. But it's also trying to delete all the other infrastructure. Both projects share the same 'dev' stack. What I am doing wrong? Or this is expected as first step to move things around? Thanks a lot

I started playing with running pulumi out of a jenkins job. One project went fine, but this new projet has an issue creating asset archives.

new pulumi.asset.AssetArchive({
      ".": new pulumi.asset.FileArchive("./lambda/CSVDownload"),
    })

which is a typescript function that creates a zip of some lambda code. When I run that out of jenkins, I get errors

Error: failed to register new resource csvDownloadLambda [aws:lambda/function:Function]: 2 UNKNOWN: failed to compute archive hash: couldn't read archive path './lambda/CSVDownload': stat ./lambda/CSVDownload: no such file or directory

Which I think is something to do with jenkins workspaces and symlinks. While this may end up being a typescript specific question (and I should head to #typescript) I'm wondering if anyone else has hit similar bumps on running out of a ci pipeline, or any suggested paths to go down for resolving this. I can't be creating a new pattern.

There is a PULUMI_CONFIG_PASSPHRASE on a stack without any secrets. How can i blank the passphrase if I do not know it?

Trying to upgrade Pulumi via Homebrew, and getting an error

Error: No such file or directory @ rb_file_s_stat - /usr/local/Cellar/pulumi/2.25.0/INSTALL_RECEIPT.json

Anyone else having the same issue?

Something I’ve noticed with the latest GA version of pulumi. With azure network resources, sometimes (very rarely but common enough to raise an eyebrow) asynchronous tasks like build a Vnet and subnets will sometimes fail at building the subnets because it doesn’t wait for the Vnet resource to complete building. On 📱please excuse typos.

👋 hello! I have a couple k8s questions: • I need to manage kubernetes clusters in multiple clouds (AWS, GCP). Would the best recommendation be to use Crosswalk for Kubernetes? • I’d like to use the newly released Autopilot in GKE. But it looks like the pulumi API doesn’t support it yet. Do you have an ETA on when it might be supported? Thanks!

Noob approach question: I want to provision an eks cluster, part of my bootstrapping process requires resources to be created just long enough to start the thing that’s going to continue running. I can’t find any good examples for temporary or ephemeral resources in pulumi. I believe the idiomatic way to do this is with a dynamic provider to make these calls. Is dynamic providers the right way to go? or is there some other mechanism?

Is there a way to mock the pulumi engine for testing?

i'm trying to destroy a test stack, in which i have marked a lot of resources as protected. how can i easily list these URNs, so i can do something like

pulumi state list | grep protected | xargs -I{} pulumi state unprotect {}

?

hi, I use pulumi install ingress on aws eks, but the pulumi always tells me：

* the Kubernetes API server reported that "yyy/xx-ingress" failed to fully initialize or become live: Resource operation was cancelled for "xx-ingress"
        * Ingress has at least one rule that does not target any Service. Field '.spec.rules[].http.paths[].backend.serviceName' may not match any active Service

Hi all, I am experiencing an error when importing a resource into pulumi. When I run

pulumi import -f import-resources.json

with the following file (abbreviated for clarity - the other resources import fine) I get an error.

{
  "resources": [
    {
      "type": "gcp:compute/vPNTunnel:VPNTunnel",
      "name": "vpn-tunnel",
      "id": "projects/<prj>/regions/<region>/vpnTunnels/vpn-tunnel"
    }
  ]
}

Here's the error:

Diagnostics:
  pulumi:pulumi:Stack (network-testing):
    error: update failed
 
  gcp:compute:VPNTunnel (vpn-tunnel):
    error: gcp:compute/vPNTunnel:VPNTunnel resource 'vpn-tunnel' has a problem: Required attribute is not set. Examine values at 'VPNTunnel.SharedSecret'.
    error: one or more inputs failed to validate

Since I did not see a documented way to set this attribute I tried creating the object in my pulumi programming with the

_import_

annotation like so:

vpn_tunnel = gcp.compute.VPNTunnel(
    "vpn-tunnel",
    name="vpn-tunnel",
    shared_secret="redacted",
    target_vpn_gateway="redacted",
    opts=pulumi.ResourceOptions(
        import_="projects/<prj>/regions/<region>/vpnTunnels/vpn-tunnel",
    ),
)

However, that resulted in an error as well (when running

pulumi preview

):

warning: inputs to import do not match the existing resource; importing this resource will fail
    = gcp:compute/vPNTunnel:VPNTunnel: (import)
        [id=projects/<prj>/regions/<region>/vpnTunnels/vpn-tunnel]
        [urn=urn:pulumi:testing::network::gcp:compute/vPNTunnel:VPNTunnel::vpn-tunnel]
        [provider=urn:pulumi:testing::network::pulumi:providers:gcp::default_4_21_0::9400fa64-3729-3efb-bb74-7183febcae7b]
      + sharedSecret: "redacted"

How can I import that resource cleanly?

I'm trying to build a pulumi stack from new with from codecommit,

pulumi new <https://git-codecommit.eu-west-1.amazonaws.com/v1/repos/pulumi_pki> --cwd .\test2\ --non-interactive

Seems easier with github as you can obtain a token, but is there a way with codecommit? Maybe passing secrets? Getting the error 

error: authentication required

. I've added AWS keys as environment variables as well as in the credentials file.

Hi, Is possible to run pulumi code using multiple cloud providers at the same time? I need to create resources in azure and aws when pulumi is executed: ej: const vnetwork = new vnet.Vnet('VirtualNet',{resourceGroup}); //for azure provider =new aws.Provider({....}) const main = new aws.ec2.Vpc("main", { //for aws cidrBlock: "10.0.0.0/16", },{provider:provider }); I have tried setting the providers but always I'm loosing the credentials: error configuring Terraform AWS Provider: no valid credential sources for Terraform AWS Provider found. If I comment the azure line it starts working.

Question about pulumi iterating over S3Object resources when deploying a static website and each file is a separate resource pulumi manages. How does pulumi compare & diff file content? Does it download and compare, or do it some other way?

Anyone knows if there's a Pulumi alternative to this? https://github.com/dealmore/terraform-aws-next-js

I've been bashing my head for two days on this and I still don't understand the issue. (https://pulumi-community.slack.com/archives/C84L4E3N1/p1618418077072900) Even if I create a completely new project (from the template) and then I do a pulumi up (new folder, etc, the only 'configured' thing is that I am logged in pulumi (S3/AWS)) the new project tries to delete all my resources that are in another project. Is there really no other way to get out of this situation? As we are not yet in production with our code I am tempted TBH to start from scratch but before I do thats I was wondering if I am doing something completely stupid with the tooling. I don't see why you can't just start as many projects as you want and keep adding state.

nice it seems like the new pulumi version removes pending operations automagically (do export/import)

What is the difference between a native provider and non native provider in Pulumi? Can you help me understand? Im being asked by leadership at my company and this is kind of the point where they decide whether or not we go with Pulumi or not as our defacto tool