Can I specify a default

aws:region

and

aws:profile

for a project, as opposed to in each stack?

hi, trying to use AWSX package with 4.24.1, is that still valid? I'm getting an error telling me that there is no local package.json in the project, but for all other of my resources I have a package.json further up my directory tree and it seems only AWSX doesn't like that configuration. I have all my node_modules and package.json in the

/data/pulumi

directory built into my container:

error: Error: Error reading file '/data/pulumi/infra/aws/projects/sitebuilder/environment/apigateway/package.json' when computing package dependencies. Error: ENOENT: no such file or directory, open '/data/pulumi/infra/aws/projects/sitebuilder/environment/apigateway/package.json'
        at readFile (/data/pulumi/node_modules/@pulumi/pulumi/runtime/closure/codePaths.js:167:19)
        at computeDependenciesDirectlyFromPackageFile (/data/pulumi/node_modules/@pulumi/pulumi/runtime/closure/codePaths.js:150:22)
        at /data/pulumi/node_modules/@pulumi/pulumi/runtime/closure/codePaths.js:115:36
        at /data/pulumi/node_modules/read-package-tree/rpt.js:162:20

This message was deleted.

hi folks, has anybody seen anything like this:

Is there a way to mark a resource to only be run once? For example, I want to start an AWS elastic beanstalk app with a sample application right away. From then on, I would like to deploy outside of pulumi. Of course I could add and remove code each time, but I’d be interested in a more automated approach. Maybe using a conditional configuration option?

hi all

how do i create a server listener rule for a second listener for my AWS NLB?

I'm trying to target an URN of a new resource to be created and pulmui is telling me that resource doesn't exist in the stack -- is it not possible to use --target with new resources?

hey all, we've been using pulumi for our kubernetes resources, and overall its pretty nice but often times we're getting "timeout" errors speaking to the GKE cluster which is strange because its GKE and sure we have like 20 resources we're deploying, but that is quite small, any suggestions?

for pulumi/kubernetes, I assume it will generate the yaml and essentially run the equivalent of kubectl -f apply behind the scenes, right? .. so I should expect pulumi up for a k8s config to work very similarly to hand written kube yml file? but of course code-generated/templated?

I am looking at how I structure our infrastructure within Pulumi. Is it right to have a single project with multiple stacks eg one stack to set up resource groups, another for networks, then for databases of various ilk and so on? The laternative is to have them in multiple projects, one for resource groups, one for networks, databases, key vaults and so on?

How can I diagnose why the GitHub action cannot edit my PR comment?

Is there a channel for the auth0 provider? Have some unusual behaviour and we're interested in some new resources that do have terraform definitions.

Hi folks, is there a way to get organizations to play well with

pulumi login --local

? If I try to do a

pulumi stack init myusername/something

I get back a

error: stack names may not contain slashes

My understanding is that

stack init

respects org names, and under file:// your org name is just your username.

Hey there does anyone have advice how to configure package/tsconfig .json files so that i can

npm install

a pulumi repo via a github url? I keep getting error

SyntaxError: Cannot use import statement outside a module

when running pulumi in the repo that installed the pulumi package. It seems i may need to set the “main” field in package.json to point to the compiled entrypoint .js file. However, that then messes up pulumi when i try and run it locally from the repo. As now pulumi tries to invoke the index.js file instead of the index.ts file. I’m trying to mimic how an installed pulumi npm module works. There are .js, .js.map, and .d.ts files. No .ts files. I’m not sure how to replicate this so that the install via github url works, as well as running pulumi within the pulumi pkg repo itself (I do that for testing) I can’t figure this out 🙂 Any advice would be much appreciated

Hey folks, I get the error

Exception: 'dependsOn' was passed a value that was not a Resource.

, but I don't explicitly set the

dependsOn

parameter. I boiled it down to one specific parameter creating the problem. I create an EKS cluster in my first project, export

cluster.core

, import it in my second project and use it as the

cluster

input to a

ManagedNodeGroup

. Currently it's the only resource created in the second project and every other input is given explicitly (= not as an

Output

type). Does any one have a clue where the error could originate or even how to solve it?

With the automation api, can I set a project to be one within an organization ? I have a devops user which is part of the organization and I setup the automation arguments as :

const stackArgs: InlineProgramArgs = {
      stackName: 'some name',
      projectName: "org/project",
      program: inlineProgram,
    };

However, when executing this i get

error: could not create stack: provided project name "org-project" doesn't match Pulumi.yaml

. However, there's no

pulumi.yaml

file within the folder I run this so I'm a bit confused. everything works fine if i don't use the

org

but that creates the project within the devops user environment. Any ideas ?

A stack config typically looks like this (ignoring structured config for the moment):

config:
  <your-project-name>:<key>: <value>

Providers often look up their config with another "project-name", e.g.:

config:
  aws:region: eu-west-1
  <your-project-name>:<key>: <value>

If I create an abstraction in a reusable library, I could for example use a separate project key to store all the config for my component.

config:
  aws:region: eu-west-1
  <your-project-name>:<key>: <value>
  <my-component-name>:<key1>: <value1>
  <my-component-name>:<key2>: <value2>

Do other Pulumi developers use this ability for such use cases? Would this be a good pattern to use, or rather an anti-pattern?

Hi, is there a best practice for running

npm build

inside pulumi? Here's the use case: We need to build and deploy a frontend REACT application to S3. And we need to know which stack to deploy while building the application, so we can set up the environment variables properly. For example, if we are deploying to

dev

, we should run

ENV=dev npm build

. Since we can easily get the stack name in pulumi, it would be convenient if I can run

npm build

inside pulumi.

Hey guys. Has anyone created a typescript pulumi project that utilizes absolute paths? I’m having many complications with using them at runtime. There are issues on Github about this. But it isn’t so simple to get absolute paths working. But am curious if anyone here is using this in their projects.

I wonder how do you guys profile your CI builds? I observe performance degradation in Github Action builds which I can't reproduce on my local.

hi all, in Terraform, we can set output to expose certain attribute of the resource of a module, so that other module can refer to it directly when creating other resources that depends on that attribute, how can I do that in Pulumi?

can anyone retrigger this job https://github.com/pulumi/pulumi-aws/actions/runs/1441710772 please ?

Hi. Can someone confirm for me whether the

new CustomResourceOptions { Protect }

option should be preventing destruction of resources by my pulumi up script or not?

Am I correct in saying it doesn't work on the first pass?

I need to protect a cluster from possible deletion, but the first pass might change the vmSize. Anyone know how to order things so it can be protected even on first pass?

Can pulumi be used for deployment of static web site content? For example, to google storage containers? Or is this something to be avoided? (I ask as there's a line some like to observe between infra management and releases.)

For storing state file of my pulumi stack I am using s3 bucket. Should I use single s3 bucket for all stacks/env? or should I create 3 different buckets for state file of 3 different stacks? Which is better/preferred? Our aim it to keep state file of different env as isolated as possible. Please suggest.

I am using the

pulumi_eks

package to set up an EKS cluster in a project. I have to import the cluster in a second project and ran into a problem:

pulumi_eks.Cluster

does not provide a

get()

method. Under the hood there is a child resource of type

pulumi_aws.eks.Cluster

with a static

get()

method but it does not work ,Pulumi can't find it. Did I miss something or is there no way of importing the cluster if using the

pulumi_eks

package to deploy it?

Is it possible to initialize a stack without creating a

.yaml

file in the current folder? (I plan on using

--config-file

)