I just don't want to generate a new password every run and overwrite the secret each time. I want to set it once, then use it going forward

Hi, someone here would know why when I run

pulumi up

in a container it reintalls the plugins each time while when I’m running the code locally I have no issue ? That’s driving me crazy. I’m using the official container

pulumi/pulumi-nodejs:latest

I'm still struggling to find a way to debug my github actions. Seems like the action gets stuck at pulumi/actions, and I can't see any logs until i abort it. I've cancelled one after 19h in the pending state, but since github had some issues yesterday, I can't say if that is normal or not when using the Pulumi actions image. Anyone else running

pulumi up

successfully via github actions? Pulumi up works locally for me.

i'm struggling to get secrets to work, specifically with a kubernetes.Provider. I've changed the input to be

kubeconfig: pulumi.secret(kubeconfigraw)

but the full plain text is still being included in

pulumi stack export

Any idea why my cloudwatch role is blank even when i specify the roleArn?

const createCloudWatchEventTargets = (rule: Output<EventRule>, pipeline: Pipeline) => {
    pulumi.all([pipeline.name, pipeline.arn, rule.name]).apply(([pipelineName, pipelineArn, ruleName]) => {
        const pipelineExecutionRole = new Role(`cwe-${pipelineName}-role`,
            {
                assumeRolePolicy: `{
                    "Version": "2012-10-17",
                    "Statement": [
                        {
                            "Effect": "Allow",
                            "Principal": {
                                "Service": "codepipeline.amazonaws.com"
                            },
                            "Action": "sts:AssumeRole"
                        }
                    ]
                }`
            });

        const pipelineExecutionPolicy = new Policy(`cwe-${pipelineName}-start-pipeline-execution`, {
            policy: `{
                "Version": "2012-10-17",
                "Statement": [
                    {
                        "Effect": "Allow",
                        "Action": [
                            "codepipeline:StartPipelineExecution"
                        ],
                        "Resource": [
                            "${pipelineArn}"
                        ]
                    }
                ]
            }`
        });

        new PolicyAttachment(`cwe-${pipelineName}-start-pipeline-execution-attachment`, {
            policyArn: pipelineExecutionPolicy.arn,
            roles: [pipelineExecutionRole]
        });

        return new aws.cloudwatch.EventTarget(`${ruleName}-target`, {
            arn: pipeline.arn,
            rule: rule.name,
            roleArn: pipelineExecutionRole.arn,
        });
    })
};

How to suppress the warning about upgrading pulumi version ? it’s totally messing up my output values…

Is there a way to get the full log of failing requests with pulumi? I have a failing stack because of lack of permission but it doesn’t tell me which api is invoking and I don’t know which role I should add. -d won’t help

Day 3 on trying to debug github actions, and I am going to try and figure out why an sts call towards us-east-1 is happening when the configuration on pulumi.com says eu-west-1. I will be creating a project from scratch and add small pieces at a time to see when it fails, but each hour spent is an hour lost.

Hello, I’m suffering

alias.0.name: required field is not set

error with this code

export const apiDnsRecord = new aws.route53.Record('WSAPIRecord', {
  name: domainName.domainName,
  type: 'A',
  zoneId,
  aliases: [
    {
      evaluateTargetHealth: true,
      name: domainName.domainNameConfiguration.targetDomainName,
      zoneId,
    },
  ],
})

so I tried domainName.domainNameConfigurationg.apply(config => …) to resolve empty targetDomainName but it didn’t work. what I found is that

export const domainName = …

clearly shows correct targetDomainName but I can’t get that. Any help/advice what i can try?

Hello all, I am playing with pulumi and I would like to know if there is a way to dump yaml for Openshift? all I see it this beta way (for K8) https://www.pulumi.com/docs/reference/pkg/nodejs/pulumi/kubernetes/#ProviderArgs-renderYamlToDirectory Can you help me?

Hi all 🙂 I am using interstack references, and i am hitting an issue where the value of the required output does not seem to be updated in the referring project. To clarify: Stack A is referencing B to get the value of the output Foo with the value "Bar". B changes the value of Foo to "Bas" but keeps the output name. When i do a "pulumi up" in the project related to A there are no changes even though Foo might be crucial to A. Destroying A and creating it again forced an update, but is not really production friendly 😛 I have tried doing a "pulumi refresh" in A but that did not update it, and it actually said that it did not update it. Any idea how to do this stack reference update of already used outputs? 🙂

hi i'm trying to figure out how to convert Outputs into their raw types (eg, StringArrayOutput -> []string) or input equivalents in golang. in the case I'm looking at, i'd want to range() over the output elements and I can't figure out how to get access to the elements.

hey, team, I’ve created Databricks pulumi provider, it works for me already, but I am not sure if we have a checklist to make pulumi it community ready https://github.com/XBeg9/pulumi-databricks

Is it possible to import an existing resource via the

import

property, without having to also specify the

name

property?

is it possible to output an attribute of a resource marked as secret as NOT secret? i.e.

export const registryCreds = pulumi.secret(new digitalocean.ContainerRegistryDockerCredentials("container-registry-creds", { registryName: registry.name }).dockerCredentials);

const registryCredsKubeSecret = new k8s.core.v1.Secret('registry-creds-kube-secret', {
  type: '<http://kubernetes.io/dockerconfigjson|kubernetes.io/dockerconfigjson>',
  metadata: {
    namespace: appsNamespaceName
  },
  data: {
    ".dockerconfigjson": registryCreds.apply(v => Buffer.from(v).toString('base64')),
  },
}, { provider })

// How do I make the output below NOT secret? 
export const registryCredsKubeSecretName = registryCredsKubeSecret.metadata.name

registryCredsKubeSecretName

is secret by "lineage", however, it's not sensitive so I want output in plain text

I don’t think you can just convert it but if you can log it from inside the apply callback and get the same sort of functionality if all you need is to see it printed. https://www.pulumi.com/docs/intro/concepts/programming-model/#how-secrets-relate-to-outputs

yeah that works, I was hoping for a built-in function that's as clean/simple as calling

pulumi.secret

to do the inverse

Hi there 👋 I’m working on a prototype for using Github Actions with Pulumi. Is it considered best practice to fill a seat with a Pulumi “machine user” in order to generate the necessary token? I’d like to decouple the token from other accounts, but since there doesnt seem to be any formal or programmatic way to set up a machine user (other than just creating a normal user via email) I thought I’d ask here

Is there a way to hide/mute the deprecation warning on kubernetes apiVersions when running

pulumi up

?

Is there something like Atlantis for Pulumi?

Hey, team. What’s going on with helm chart hooks? Does pulumi support it? I’ve faced the issue, because helm chart uses the hook annotations and they are not applied correctly. Thanks

Hi guys, I'm using pulumi tags to organise a little better my stacks. Right now I'm using the commands I found here: https://www.pulumi.com/docs/reference/cli/pulumi_stack_tag/. However, it is a little hard to know which tags your stack is using if you don't go to pulumi page. Is there a way to define tags in stack yaml? I think it would be great to support this, because tags would be easier to maintain and the visibility would improve.

Has anyone else experienced pulumi getting stuck on 'pulumi up'? The cli basically runs forever, doing apparently nothing

I have a pulumi stack where I create a docker image and push it to my registry. On my local machine it doesn't push unless the image is changed but on Github Actions it builds and pushes every time. Is there any way I can cache those builds?

Is it possible to change the location of this file? https://www.pulumi.com/docs/guides/continuous-delivery/github-actions/#branch-mappings

Hello !!

Is it possible to configure pulumi to work via a web proxy ?

I've got a weird thing I've never seen before where

pulumi destroy --non-interactive --yes

is just emptying the state file instantly without making any changes to the resources themselves