Notice anything? 🙂 https://resource-types.concourse-ci.org/

how can i submit a PR? the readme says to run

make travis_pull_request

but it just outputs

moved to GitHub Actions

Wanted to get some opinions before I file a Github issue/pull-request. The S3 state backend is very slow in my opinion. When compared to Terraform, Pulumi takes much longer to perform updates for 100+ resources using S3 as the backend for the state. I think this is because Pulumi checkpoints the state file to S3 after every action whereas Terraform waits and updates the state after it has finished performing all of its actions. Terraform also locks the state file before doing anything while Pulumi does not, so this might be a blocker to mimicking Terraforms state management behavior. Am I on the right track?

Hi everyone! I’m trying to make the pulumi-tf-provider-boilerplate easier to use by updating README, fixing Makefile, etc. Please let me know what was hard for you when you tried to use it for the first time. What went wrong? What was unclear? That’s my first PR on that topic. Maybe you would have some comments on that too.

Wanted to make a small blog on how to use ChartMuseum to mirror Helm repositories, it turned out to be 99% on how to deploy the entire infra with Pulumi, anyway, maybe someone use it to get started: https://vanderveer.be/2021/chartmuseum-mirror/

i wrote this dynamic provider so we can build policies dynamically as we add resources that share a KMS key. is this a bad idea? 😅

#azure How long should

make build

take for azure-native? I only ask because I span up a GitHub Codespace and ran that command (after

make ensure

) and then the server became totally unstable, twice. And when I say unstable, I mean that I cannot connect to the Codespace again via the web or VS Code. I'm wondering if it is just taking a lot of resources to build or something. Really strange.

@lemon-agent-27707 @tall-librarian-49374 looking for thoughts on something. Currently I am wiring up the Pulumi Automation preview to our internal CD app and we're looking to preserve that pulumi log output for historical purposes. So we'll probably hook into that

OnOutput

delegate to capture those logs. But currently if you look here you'll see that the

OnOutput

delegate is only invoked when the StdOut stream is written to and not when StdErr is written to. I believe I just mimicked the behavior from the other SDKs when I did that.. but that means that if their is an error it will only be reflected in the resulting exception and not in our historical logs that we built from the

OnOutput

delegate. Do you think that maybe we should have a second delegate for

OnError

, or maybe

OnOutput

should be invoked for both?

Is there a general reference for getting setup locally to develop on pulumi packages, specifically awsx in this case? I’ve been reverse engineering the process from the

Makefile

, but would like to avoid blindspots. I found that

yarn

and https://github.com/pulumi/pulumictl are required, then I can run

make build_nodejs

as long as I have

tsc

in my

PATH

, so I’m able to run

make test

. Additionally, I’d like to be able to test my changes to this package in my Pulumi project before opening a PR.

Hey, does Pulumi allow people to build providers and submit them to be added to the tool?

Hi! I raised a PR for a minor documentation edit. Is there anything else I need to do or be aware of? For context, it’s to address this: https://pulumi-community.slack.com/archives/CRH5ENVDX/p1617285357144400

Hi everyone I ran into this issue when building. I currently have:

go 1.16.2
dotnet-core 3.1
dotnet-runtime 3.1
aspnet-runtime 3.1
yarn
nodeJS 15.12.0
python 3.9.2

Can I get some eyes on this? I would love for this to get merged so I can delete the stop-gap code from my project: https://github.com/pulumi/pulumi-awsx/pull/659

I promise I will get it reviewed and tested today @millions-furniture-75402 - my apologies here, I've been busy with the creation of the 3.0-beta.x releases

Hello, I am very new to all of these. The break points I am adding to the tests are not working, they are just ignored. This is not happening with go programs, so it is may be not the ide. I was wondering if somebody could help me. Also I am not sure if this is the right channel, if there is another channel for people who are just new, I would be happy if someone can refer them to me

I've got a vs code dev container working for a builds on windows via wsl2... https://github.com/pulumi/pulumi/issues/4963#issuecomment-827312020

I can

make ensure

then

make only_build

but then

make

fails to run all the tests

I don’t know if you know about this: https://nubesgen.com it is a generator to generate terraform code for azure by selecting some stuff in a UI. I went a head and placed a first draft to support pulumi - but I don’t know if it gets accepted… https://github.com/microsoft/NubesGen/pull/117 WDYT?

Hi, I am trying to understand the pulumi source code. What have tried a couple of example of pulumi for kubernetes and read the architecture. I am reading the code , but what would be really helpful is that, if I could debug the flow. meaning if I want to see what package or files are being executed at each step inside pulumis source code. Could someone help me with this?

I tried this https://github.com/pulumi/pulumi/issues/1372 so far but was unsuccessful.

Hi all! We started adoping Pulumi in our organisation but we found the user experience could be improved. We have a specific issue that only applies if one uses S3/Azure blob/GCP storage to store states. At the moment, we've got about 50 AWS accounts and several Azure subscriptions that we work on, and it's a pain in the ass to do

pulumi login

whenever we want to switch between projects, especially if working on several at the same time. I found that you can specify

backend:
  url: <s3://bucket>

in

Pulumi.yaml

, but it still requires setting

AWS_PROFILE

env var to get the right profile. The same story is with azblob:// backend but it's even worse 😄 Have to set env vars with storage account credentials. And also can only specify container in the URL. We've got a workaround using direnv but it's not ideal. However! I figured out the limitation was in https://gocloud.dev/ library, as it didn't allow setting profile in the URL. But now it does for s3: https://github.com/google/go-cloud/issues/2865 Pulumi uses a forked version of gcloud.dev, so if there's no strict reason to not update, it would be nice to get the latest. I've got a POC for azure blob as well, that would allow storage account name and possibly subscription id (depends on the underlying library) in the URL.

Hi @fierce-ability-58936 - we will pick up the new release when it is made!

thanks for pointing that out - that's an excellent addition

Hello. Me and my team are trying to contribute to Pulumi by adding JVM support. I have question - how the language specific classes are being generated in 

sdk/

  folder? I wrote minimal Java code generator (

gen.go

) which generates 

Config.java

 file. Also, I discovered that 

GeneratePackage()

 function is executed only in tests and I guess this is “main entrypoint” for generating files, right? But I found out that code is being generated inside the internal testdata folder. Could you give us some information, how Pulumi team generates language files in 

sdk/

 folder?

Hi everybody, I wanted to start working on this issue, but having a hard time understanding how development works, probably because of the fact that I’ve never done Go before and I’m a complete noob (TS/JS is my main language). I’ve managed to fork the project and install dependencies, I can even build using the

make

command (although some errors do pop up when tests are run)… But what I don’t understand is … once I make the change in the CLI code, how can I actually see my changes? For example, let’s say I wanted to add a simple

--path

parameter to the CLI’s

watch

command, is there a way for me to run the CLI and see it in action immediately, once I’ve done the code change? Sorry for the noob questions, and I’m sorry if I missed a piece of info somewhere in your contributions guide. 🙂 It’s not that I’m asking a step-by-step solution / answer. If there’s a URL that you might find useful / that could get me up to speed faster, I’d appreciate that too. 🙂

Hi everyone! I’ve built a proof of concept of using Pulumi Automation API in k8s operator. Is this example good enough to publish it and send a PR with a link to pulumi automation examples? The main idea of the operator is to give easy access to cloud resources for developers. For example, developers work on an authorization service and they need a Redis cluster. They just create a small CRD in the service’s helm chart and apply it. Something like this:

kind: RedisCluster
metadata:
  name: auth-redis
spec:
  memory: 16Gb

k8s operator creates Redis Cluster in AWS with all companies security policies and best practices while developers don’t even need to know in which cloud their code runs. Proof of concept is written in Go with OperatorSDK.

hey i have this PR to fix a few comments in pulumi-newrelic https://github.com/pulumi/pulumi-newrelic/pull/204

Hi together, I’ve migrated our whole infrastructure setup from Terrafom to Pulumi. Pulumi is a very great tool and the whole company supports the migration. We use AWS and Heroku. For Heroku I’ve created the pulumi-heroku provider with the

pulumi-tf-provider-boilerplate

repository a time ago to start and migrate. (https://github.com/moneymeets/pulumi-heroku) Now I would like to upgrade to the new pulumi version 3.x and I need to upgrade the provider. I’ve seen that there are a lot of changes and need some help. I would like to share the provider with the community and would kindly ask if there is somebody who can help and perhaps adopt this provider? Thanks in advance. Best wishes Marcel