pulumi-community #general

general

colossal-australia-65039

02/09/2021, 12:02 AM

I'm attempting to use

StackReference

to the

test

stack with everything using the same S3 backend but I run into this error

pulumi:pulumi:StackReference (test):
    error: Preview failed: decrypting secret value: failed to decrypt: incorrect passphrase, please set PULUMI_CONFIG_PASSPHRASE to the correct passphrase

I can't find documentation on how to resolve this. I've tried exporting

PULUMI_CONFIG_PASSPHRASE

as the one for

test

instead of my current stack, but then i get another error since the secrets password for the current stack is mismatching

gifted-terabyte-92288

02/09/2021, 1:17 AM

👋 I have somehow completely hosed my local Pulumi CLI (Mac OS 10.15.7 | Pulumi CLI v2.20.0). Whenever I run any command like

pulumi stack select

pulumi refresh

I get the following:

error: invalid character '"' after top-level value

Some commands work though - like

pulumi stack ls

(but I can't then switch to any of the stacks). I have tried uninstall and re-installing the cli (

brew remove pulumi

brew install pulumi

), but between installations it remembers who I am. I'm assuming there's some cached configuration that is corrupted - any idea where that file may be (I checked Homebrew cache)?

ambitious-father-68746

02/09/2021, 2:04 AM

I'm having trouble with my Pulumi runs taking a long time to complete. I have roughly 1200 resources on my stack and running

pulumi pre

is over in a minute, but the smallest change with

pulumi up

, like updating just one resource, doesn't matter which, takes more than 30 minutes to run. This happens even if I use

-t

. My backend is AWS. I've enabled some verbose logging and there is a lot of

incoming output value translated

and

incoming output property translated

, it seems to be covering the whole of the stack, once again even with

-t

. It ends with a

waiting for quiescence

with a counter on the upper hundreds going to zero, which I can't seem to be able to even find on Google. It almost feels like the whole of the stack is being processed even though there is a single change to be made. I'm not sure how to progress from here, having to wait between 30 minutes to an 1+ hour just to add an environment variable to a service is killing us. Any ideas? Thank you.

wide-crayon-4093

02/09/2021, 3:46 AM

are there any examples of pulumi ts projects that are not contained in a single

index.ts

? super excited to be moving my infra to code. for now I'm just procedurally dumping my resources in a single file, but want to eventually structure and modularize it properly. what I'm most interested in is passing dependencies and context around. (e.g k8s provider)

echoing-activity-32278

02/09/2021, 11:15 AM

I am thinking of laying out the directory as follows: infra/ dev/ stg/ prod/ In this sense, naming the stack as dev is meaningless as it contains both the stg and prod setup. So I wonder what’s the desired workflow for pulumi?

narrow-battery-9106

02/09/2021, 12:21 PM

Hi guys, did some one tried using crd2pulumi with tekton crds? The generated code seems to be lacking spec property for all resources I am generating it from there https://storage.googleapis.com/tekton-releases/pipeline/latest/release.notags.yaml

spec:
  group: tekton.dev
  preserveUnknownFields: false
  versions:
    - &version
      name: v1alpha1
      served: true
      storage: false
      # Opt into the status subresource so metadata.generation
      # starts to increment
      subresources:
        status: {}
      schema:
        openAPIV3Schema:
          type: object
          # One can use x-kubernetes-preserve-unknown-fields: true
          # at the root of the schema (and inside any properties, additionalProperties)
          # to get the traditional CRD behaviour that nothing is pruned, despite
          # setting spec.preserveUnknownProperties: false.
          #
          # See <https://kubernetes.io/blog/2019/06/20/crd-structural-schema/>
          # See issue: <https://github.com/knative/serving/issues/912>
          x-kubernetes-preserve-unknown-fields: true
    - !!merge <<: *version
      name: v1beta1
      storage: true
  names:
    kind: Pipeline
    plural: pipelines
    categories:
      - tekton
      - tekton-pipelines
  scope: Namespaced
  conversion:
    strategy: Webhook
    webhook:
      conversionReviewVersions: ["v1beta1"]
      clientConfig:
        service:
          name: tekton-pipelines-webhook
          namespace: tekton-pipelines

I presume problem is with !!merge and &version thing but I am not sure

narrow-battery-9106

02/09/2021, 12:50 PM

it is not the problem I tried equivalent without these tags

big-potato-91793

02/09/2021, 4:10 PM

https://pulumi-community.slack.com/archives/CRH5ENVDX/p1612886121200400

big-potato-91793

02/09/2021, 4:10 PM

Does somone have an idea?

icy-jordan-58549

02/09/2021, 4:23 PM

can’t find this in the docs, would be great if someone can point me into it, how I can understand if

pulumi

is currently running in

preview

mode ?

glamorous-van-94517

02/09/2021, 5:05 PM

👋 I’m here! What’d I miss?

🎉 1

clever-lawyer-94920

02/09/2021, 5:39 PM

Is Pulumi only for infra provisioning or can it be used as a configuration management tool as well? For example it is recommended terraform be used as a provisioning tool along with ansible as the config managment tool. Is it the same for Pulumi?

average-school-38756

02/09/2021, 9:08 PM

after i import an existing resource into a project, why do i have to keep the resource name hard-coded? if i remove it, Pulumi wants to do delete-replace just to give it a new name.

broad-dog-22463

02/09/2021, 11:20 PM

message has been deleted

red-football-97286

02/10/2021, 11:44 AM

Hi @broad-dog-22463 I think provider then language would read well.

broad-dog-22463

02/10/2021, 11:45 AM

Hi @red-football-97286 if you can, can you make a wee comment on the GH issue? We want to try and collate the responses

red-football-97286

02/10/2021, 11:45 AM

Sure 🙂

quiet-lock-69596

02/10/2021, 12:51 PM

Indeed in favor of @red-football-97286s alternative. Did a thumbs up for that on the GH issue 🙂

🙌 1

broad-dog-22463

02/10/2021, 1:04 PM

thanks @quiet-lock-69596!

gorgeous-beach-44078

02/10/2021, 2:00 PM

Heya guys, Having some issues with pulumi which I hope one of you will be able to assist with. I am trying to use an application called "prisma.io" to generate database and graphQL scheme's automatically and handle how the application makes the calls to the database. However I encounter an error whenever I try to upload the code due to

@prisma/client

not initialising. A simular issue is found within

serverless

, where the main issue was webpack running another yarn install prior to packaging, which was not running my prisma generate properly. This caused my @prisma/client to be either empty, or at least not include the required binaries. The solution when using

serverless

would be to create an custom webpack. How would I solve the same issue in pulumi?

wide-crayon-4093

02/10/2021, 2:58 PM

hey guys. how can I use

default

namespace from k8s? I did an import, but if I delete the k8s cluster it will try to remove the default namespace first, which is not possible.

wide-crayon-4093

02/10/2021, 3:01 PM

I'm realizing that using the default namespace may not be a good idea

bitter-application-91815

02/10/2021, 4:08 PM

hey guys, I never provided a passphrase when creating a stack, now that i have created an s3 accesskey and secret, the secret is returned encrypted. Is there a way to decrypt that secret

bitter-application-91815

02/10/2021, 4:08 PM

or do i need tear down the stack, recreate with passphrase

fast-oyster-42706

02/10/2021, 4:13 PM

Doing a Pulumi POC and ran into what I think is a bug but trying to confirm expected behavior. After stack was deployed I manually deleted the resource group in azure. I expected Pulumi to reconcile the missing rg and recreate it. What is the expected behavior if a stack resource is deleted manually?

icy-jordan-58549

02/10/2021, 4:18 PM

The Pulumi runtime detected that 50 promises were still active
    at the time that the process exited. There are a few ways that this can occur:
      * Not using `await` or `.then` on a Promise returned from a Pulumi API
      * Introducing a cyclic dependency between two Pulumi Resources
      * A bug in the Pulumi Runtime
    Leaving promises active is probably not what you want. If you are unsure about
    why you are seeing this message, re-run your program with the `PULUMI_DEBUG_PROMISE_LEAKS`
    environment variable. The Pulumi runtime will then print out additional
    debug information about the leaked promises.

tall-shoe-66375

02/10/2021, 6:40 PM

Hello, I have some questions. which is the channel or procedure to do this ?

fast-oyster-42706

02/10/2021, 9:37 PM

Having an issue with unit test due missing “missing required configuration variable ‘sa:accounttier’“. Read some posts around this and the testing docs but unable to make it work. What is the best way to handle config for unit testing with mock. I am using Golang.

ambitious-appointment-66307

02/10/2021, 10:18 PM

Hi team - I just upgraded my pulumi version and tried to make a change to my stack (first for a couple of months) and I’m getting this error

error: failed to discover plugin requirements: exit status 1

ambitious-father-68746

02/11/2021, 3:18 AM

Hi, any reason why an empty

pulumi up

with no changes to the stack (containing ~1200 resources) would take ~30 minutes to run?

pulumi pre

only takes 1 minute and a half. The logs show that Pulumi is never stuck and there's a whole bunch of

registering resource

preparing resource for RPC

and other messages on the logs that feel like Pulumi is doing stuff to all resources of the stack, which makes no sense. I'm using Python and mostly AWS. The final output of

pului up

shows:

Resources:
    1176 unchanged

Duration: 28m1s

Title

ambitious-father-68746

02/11/2021, 3:18 AM

Hi, any reason why an empty

pulumi up

with no changes to the stack (containing ~1200 resources) would take ~30 minutes to run?

pulumi pre

only takes 1 minute and a half. The logs show that Pulumi is never stuck and there's a whole bunch of

registering resource

preparing resource for RPC

and other messages on the logs that feel like Pulumi is doing stuff to all resources of the stack, which makes no sense. I'm using Python and mostly AWS. The final output of

pului up

shows:

Resources:
    1176 unchanged

Duration: 28m1s

View count: 1