edit: nevermind, found out how and posted as a comment in this thread for the pulumi postgres package, is there a way to set the database host, username, etc. through code WITHOUT having to use environment variables or pulumi.yml configs? For example with Typescript I tried

import * as pg from "@pulumi/postgresql"

pg.config.host = 'redacted'
pg.config.username = 'redacted'
pg.config.password = 'redacted'

const serverRole = new pg.Role('redacted', {
    login: true,
    password: 'redacted',
    name: 'redacted',
})

but I'm getting the error

error: Error initializing PostgreSQL client: error detecting capabilities: error PostgreSQL version: dial tcp :5432: connect: connection refused

The same creds work when I export as an env var.

Quick question, I’m trying to replicate what your ec2 depoyment from terraform to pulumi. I was having a question about this property that we have inside terraform

lifecycle {
    create_before_destroy = true
  }

this is something I can do also in pulumi?

Is there any way to debug this :

Type                 Name              Plan     Info
     pulumi:pulumi:Stack  services-staging           1 error
 
Diagnostics:
  pulumi:pulumi:Stack (services-staging):
    error: an unhandled error occurred: Program exited with non-zero exit code: -1

on a

pulumi up

with no stack trace and no obvious way to understand the problem, appearing suddenly without any change to the codebase ? (TS flavor) Even tried to checkout some old commit ref, still happening..

Hi! 👋 I was a bit curious about

Output.all()

, is there any way to access the outputs one wants in the lambda by name, instead of index? It quickly gets confusing with a more substantial piece of code like this one:

environment = Output.all(
    databases["airflow"].address,  # 0
    databases["airflow"].password, # 1
    databases["grafana"].address,  # 2
    databases["grafana"].password, # 3
    databases["redata"].address,   # 4
    databases["redata"].password,  # 5
    base_url,                      # 6
    sd_namespace.name,             # 7
).apply(
    lambda args: [
        # Airflow DB
        {"name": "AIRFLOW_CONN_METADATA_DB", "value": f"<postgres+psycopg2://airflow:{args[1]}@{args[0]}:5432/airflow>"},
        {"name": "AIRFLOW_VAR__METADATA_DB_SCHEMA", "value": "airflow"},
        # Airflow Config
        {"name": "AIRFLOW__CORE__LOAD_DEFAULT_CONNECTIONS", "value": "False"},
        {"name": "AIRFLOW__CORE__SQL_ALCHEMY_CONN", "value": f"<postgres+psycopg2://airflow:{args[1]}@{args[0]}:5432/airflow>"},
...

As you can see, I've taken to labeling each output with its index just to not get too lost.. 😅

Hi when running pulumi up i see the docker debug on each run, anyone know how i disable the output?

pulumi:pulumi:Stack (xxxxxxx):
    #1 [internal] load build definition from Dockerfile
    #1 sha256:xxxxx
    #1 transferring dockerfile: 237B done
    #1 DONE 0.0s
    #2 [internal] load .dockerignore
    #2 sha256:xxxxx    #2 transferring context: 2B done
    #2 DONE 0.0s

when using the

awskms

secrets provider, what format/encoding is the ciphertext stored in? I get something like

"ciphertext": "v1:NRtBgOKsQ3V/dF57:LHwdjA….

in my stack outputs which is not base64. (I want to decrypt it and get the plaintext)

Hello 🙂

Hi, a warm hi from 🇨🇴

Hello everyone

hello new users! feel free to ask in #workshops if you have questions about the earlier workshop!

Curious if anyone here has any tips/advice for how they're managing multiple pulumi projects inside of monorepos with Typescript. I've run into a ton of dead-ends when it comes to using

yarn workspaces

(there are quite a few open GH issues about this). Are people here mostly just having separate repos for different projects? Or are you keeping a top level package.json/tsconfig.json and then having individual projects in different directories?

hi all, is there a way to upload files to S3 in a way that those files won't get deleted when the stack is deleted? our use case is a code-split react app and we need to keep files from older deployments around for a little longer. is

protected: true

the way to go?

hi, is there an easy way to generate yaml file after pulumi import? i thought of doing that through

pulumi stack --show-ids

but wondering if there is a tool or anything else easier

If I have the class below and want to call/use the registered output in another module, how can i call it? When i try importing the class and calling it as

pki_intermediate_environment_backend.pki_intermediate_environment_backend_path

it throws the following AttributeError:

'OLVaultPKIIntermediateEnvBackend' object has no attribute 'pki_intermediate_environment_backend_path'

Here’s the class:

class OLVaultPKIIntermediateEnvBackend(ComponentResource):

    def __init__(
        self,
        backend_config: OLVaultPKIIntermediateEnvBackendConfig,
        opts: ResourceOptions = None,
    ):
        super().__init__(
            "ol:services:Vault:PKI:IntermediateEnvBackendConfig",
            backend_config.environment_name,
            None,
            opts,
        )

        resource_options = ResourceOptions(parent=self).merge(opts)  # type: ignore

        self.pki_intermediate_environment_backend = Mount(
            f"pki-intermediate-{backend_config.environment_name}",
            opts=resource_options,
            path=f"pki-intermediate-{backend_config.environment_name}",
            type="pki",
            description=f"Backend to create certs for pki-intermediate-{backend_config.environment_name} backends",
            max_lease_ttl_seconds=backend_config.max_ttl,
            default_lease_ttl_seconds=backend_config.default_ttl,
        )

self.register_outputs(
            {
                "pki_intermediate_environment_backend_path": self.pki_intermediate_environment_backend.id,
            }
        )

I've got this warning in the Pulumi console, anyone know what it means? "uncommited changes were present at the time of this update"

hey!

when i store state files on local file system, the command

pulumi stack ls -p myproject

doesn't filter, instead list all stacks.

does any of You know good Alicloud tutorial for Pulumi?

I am looking for basic walk through setting backend on OSS + "talking" with Terraform state

Question: I’m looking (dreaming) to eventually one day setup a multi cloud azure + google + aws. Azure recently released to preview a cool tool to assist with management called Arc, is this in the pipeline for future features?

I don’t know much about Pulumi yet, but for the azure implementation why add the suffixes to the name? Couldn’t this suffix be swapped out for a tag given azure provides a good alternative to name suffixes?

Is it just a trying to keep consistent accross providers kinda solution?

Is it possible to have a resource that is "owned" by multiple stacks? My current use case is that I've got a project that creates a number of VPCs, and I've got stacks for QA and Production environments. Now I want to add a VPC that is peered to both QA and Production locations so that there will only be a single instance of it. Would it be possible/advisable to do something like create the resource in the QA stack, and then import the same resource in the production stack?

The alternative is that I create a separate project to handle creating the singleton environment and manage it as a separate concern, though I'm trying to avoid that if it's feasible.

Would there be any other ways to approach this problem? 😁

I’ve just noticed that the resource group created by azures KubernetesCluster.NodeResourceGroup doesn’t have a suffix, will this break the suffixes benefits?

Hello, I’m discussing refactoring our Pulumi deployment. We currently don’t employ CI / CD and are using a monorepo. We’re discussing breaking up our monorepo into two repos. One for SDK like Component Resources, the other for stacks which utilize Component and Custom resources.