Does anyone have an example of using a StackReference in Go?

Following up on my previous question regarding StackReferences in Go, I'm having an issue referencing an array in another stack. I have an array of AWS subnet IDs exported from an infrastructure stack like this:

ctx.Export("privSubnetIds", pulumi.StringArray(privSubnetIds))

In another stack that builds on top of that infrastructure stack, I reference the array like this:

infra, err := pulumi.NewStackReference(ctx, "org/project/stack", nil)
privSubnets := infra.GetOutput(pulumi.StringArray("privSubnetIds))

I haven't tried to run this yet, because

gopls

inside VS Code is showing errors on the

privSubnets

assignment line & when I try to reference that value inside an

ec2.NewInstance

stanza. I think the error is in the assignment line, but I'm not quite sure how it needs to be changed to work. Any suggestions?

Is there any sample Go code that shows how to setup and exec a GCP compute instance provisioner?

I'm continuing to struggle with stack references. Following up on this thread (https://pulumi-community.slack.com/archives/CCWP5TJ5U/p1595004364263000), I've adopted the recommended

ApplyStringArray

and

ApplyString

functions to the outputs from the stack reference, but trying to create an instance using a security group and subnet from the stack reference fails. I guess it's defaulting to the first subnet in the default VPC and not using my subnet at all. Code in thread below.

the code that is generated from this using tf2pulummi

resource "okta_policy_mfa" "blue-ocean" {
				name            = "Blue Ocean"
				status          = "ACTIVE"
				description     = "Blue Ocean"
				groups_included = ["abc"]
				priority        = okta_policy_mfa.svc.priority + 1
			  }
			  resource "okta_policy_mfa" "svc" {
				name            = "Service Accounts"
				status          = "ACTIVE"
				description     = ""
				groups_included = ["abc"]
				priority        = 1
			  }

looks like this

import (
        	"<http://github.com/pulumi/pulumi-okta/sdk/v2/go/okta/policy|github.com/pulumi/pulumi-okta/sdk/v2/go/okta/policy>"
        	"<http://github.com/pulumi/pulumi/sdk/v2/go/pulumi|github.com/pulumi/pulumi/sdk/v2/go/pulumi>"
        )
        
        func main() {
        	pulumi.Run(func(ctx *pulumi.Context) error {
        		svc, err := policy.NewMfa(ctx, "svc", &policy.MfaArgs{
        			Name:        pulumi.String("Service Accounts"),
        			Status:      pulumi.String("ACTIVE"),
        			Description: pulumi.String(""),
        			GroupsIncludeds: pulumi.StringArray{
        				pulumi.String("abc"),
        			},
        			Priority: <http://pulumi.Int|pulumi.Int>(1),
        		})
        		if err != nil {
        			return err
        		}
        		_, err = policy.NewMfa(ctx, "blue_ocean", &policy.MfaArgs{
        			Name:        pulumi.String("Blue Ocean"),
        			Status:      pulumi.String("ACTIVE"),
        			Description: pulumi.String("Blue Ocean"),
        			GroupsIncludeds: pulumi.StringArray{
        				pulumi.String("abc"),
        			},
        			Priority: <http://pulumi.Int|pulumi.Int>(svc.Priority.ApplyT(func(priority int) (float64, error) {
        				return priority + 1, nil
        			}).(pulumi.Float64Output)),
        		})
        		if err != nil {
        			return err
        		}
        		return nil
        	})
        }

running this code results in this

pulumi_okta [pulumi_okta.test]
/Users/jjayaweera/go/src/pulumi_okta/some_test.go:143:24: cannot convert svc.Priority.OutputState.ApplyT(func literal).(pulumi.Float64Output) (type pulumi.Float64Output) to type <http://pulumi.Int|pulumi.Int>
/Users/jjayaweera/go/src/pulumi_okta/some_test.go:144:21: cannot use priority + 1 (type int) as type float64 in return argument
FAIL	pulumi_okta [build failed]
FAIL

What is the correct syntax for this

Priority: <http://pulumi.Int|pulumi.Int>(svc.Priority.ApplyT(func(priority int) (float64, error) {
        				return priority + 1, nil
        			}).(pulumi.Float64Output)),

If I have an array of AZ names created like this:

azNames := make([]string, numOfAZs)
for idx := 0; idx < numOfAZs; idx++ {
	azNames[idx] = rawAzInfo.Names[idx]
}

where

rawAzInfo

is the result of the

aws.GetAvailabilityZones

resource, how can I use that array as an input to the AvailabilityZones argument of the

elb.NewLoadBalancer

resource? Alternately, if I have an array of instance IDs, how could I use that array of instance IDs as an input to the Instances argument of the

elb.NewLoadBalancer

resource?

bleh the AWS SDK changed Tags form a

pulumi.MapInput

to a

pulumi.StringMapInput

forcing me to refactor my code.. not huge, but an unexpected breaking change for a point release

Anyone have an example of creating an AWS classic ELB using Go? I haven't been able to find one yet.

I’m translating some of this https://github.com/pulumi/examples/blob/master/aws-ts-static-website/index.ts to Go. Performing the ACM certificate domain validation seems problematic though. In typescript it’s just this:

const certificateValidationDomain = new aws.route53.Record(`${config.targetDomain}-validation`, {
        name: certificate.domainValidationOptions[0].resourceRecordName,
        zoneId: hostedZoneId,
        type: certificate.domainValidationOptions[0].resourceRecordType,
        records: [certificate.domainValidationOptions[0].resourceRecordValue],
        ttl: tenMinutes,
    });

Translating that directly to Go gives this

certValidationDomain, err := route53.NewRecord(ctx, domain+"-validation", &route53.RecordArgs{                                                                                                                     
            Name:    cert.DomainValidationOptions.Index(<http://pulumi.Int|pulumi.Int>(0)).ResourceRecordName(),                                                                                                                               
            Type:    cert.DomainValidationOptions.Index(<http://pulumi.Int|pulumi.Int>(0)).ResourceRecordType(),                                                                                                                               
            Records: pulumi.StringArray{cert.DomainValidationOptions.Index(<http://pulumi.Int|pulumi.Int>(0)).ResourceRecordValue()},                                                                                                          
            ZoneId:  zone.ID(),                                                                                                                                                                                            
            Ttl:     <http://pulumi.Int|pulumi.Int>(600),                                                                                                                                                                                      
        })

however, that gives this kind of error

/main.go:87:4: cannot use cert.DomainValidationOptions.Index(<http://pulumi.Int|pulumi.Int>(0)).ResourceRecordName() (type pulumi.StringPtrOutput) as type pulumi.StringInput in field value:
        pulumi.StringPtrOutput does not implement pulumi.StringInput (missing ToStringOutput method)

i guess i’ll work around using

ApplyT

but feels like it shouldn’t be necessary

HI who can help to create "LoadBalancer" i have issue in params Cannot convert expression of type '[]string' to type 'String' Golang

type (
	S1LBConfig struct {
		LoadBalancer struct {
			Rules []struct {
				Lb struct {
					SecurityGroups           []string
					Subnets                  []string
				} 
			} 
		} 
	}
)

SecurityGroups:           pulumi.StringArray{pulumi.String(s1LoadBalancerConfig.LoadBalancer.Rules[k].Lb.SecurityGroups)},

Cannot convert expression of type '[]string' to type 'String'

Hey, a few questions: 1. Is there AWS EKS managed node group support in go? https://github.com/pulumi/pulumi-eks is nice but only typescript. I only see classic node groups supported in https://github.com/pulumi/pulumi-aws in the go sdk 2. Similarly, https://github.com/pulumi/pulumi-awsx does not have go support. Is there a timeline for that? I am specifically looking at the vpc best practices creation function. It would be nice to have that single function call. 3. For getting values out of a created vpc resource, I see that direct value references do not seem to be supported in Go. It looks like I should be using the Apply() function as described here. Can you tell me how that would specifically work for getting a vpc id as a string? I am looking at this example

url := vpc.DnsName.ApplyString(func(dnsName string) string {
    return "https://" + dnsName
})

Thank you

Hi, Is it possible in Go to execute a function at runtime? I would like to create a resource, write some information of a resource to a local file, then package it up and use that package in another to be created resource. I see something like that is possible in Javascript with callbacks, but can't find anything for Go. Thanks

Another question: After creating subnets my functions returns []*ec2.Subnet, but in ElasticBeanstalk I need to add them to EnvironmentSettingArray with EnvironmentSettingArgs {Name: pulumi.String("Subnets"), NampeSpace: pulumi.String("awsec2vpc"), Value: pulumi.String("subnetid1,subnetid2,subnetid3")} However how can I create a comma seperated string from the array, as when I do fmt.Sprintf("%v,%v,%v",subnet[0].ID(),subnet[1].ID(),subnet[2].ID()) I only get the pointers, not the real value as this is filled at deployment time and not after creation.

Hi guys, I'm playing around in pulumi. Was trying to have custom keyspace so that I can have a template for creating vms But getting following error Pulumi.dev.yaml

config:
  vm:instance_spec:
  - count: 1
    data_disk_size: 50
    name: cassandra
    resource_group: pu_demo
    type: Standard_A8_v2
    user_name: deployer

main.go

// vm config struct
type Data struct {
	Data []struct {
		Name              string `json:"name"`
		Type              string `json:"type"`
		Count             int    `json:"count"`
		DataDiskSize      int    `json:"data_disk_size"`
		UserName          string `json:"user_name"`
		ResourceGroupName string `json:"resource_group"`
	} `json:"instance_spec"`
}
func main() {
	pulumi.Run(func(ctx *pulumi.Context) error {
		vmConfig := config.New(ctx, "vm")
		var d Data
		if err := vmConfig.GetObject("instance_spec", &d); err != nil {
			return err
		})
		fmt.Printf("%v\n", d)
return nil
)}
}

and I'm getting this error

Diagnostics:
  pulumi:pulumi:Stack (demo-dev):
    error: program failed: 1 error occurred:
        * json: cannot unmarshal array into Go value of type main.Data
    exit status 1
    error: an unhandled error occurred: program exited with non-zero exit code: 1

Hi, I’m trying to increase testing coverage and want to know how y’all recommend mocking out lookup functions like

ec2.GetSubnetIds()

? Maybe testing these failure scenarios is only really possible using something like the integration testing utilities?

Anyone here using the datadog provider in golang? I am seeing it eat up 17GB of memory when it compiles

my coworker tracked it down to what we believe is an issue with a 4MB source file… but has anyone else seen this?

https://github.com/pulumi/pulumi-datadog/issues/68 is the issue he opened

I am looking for more complex Golang examples… Can anyone link examples for creating Secrets and (ok, that should be straight forward, but not covered by docs/examples) and for a more detailed GKE configuration?

creating secrets in secrets manager? or creating pulumi secrets?

Secrets in K8S via Pulumi

The basic structure for how to create a secret seems covered in the reference docs at least https://www.pulumi.com/docs/reference/pkg/kubernetes/core/v1/secret/

i want to create a postgres cluster, and then reuse that pg cluster across stacks this requires that the creator stack exports info on how to connect to the cluster when it comes to the provider, this is easy-- the consumer stack can just use

pulumi.NewStackReference(...).GetProvider("pg-provider-id")

however, the consumer stack also needs raw connection info (host, port, etc) in order to inject that info into kubernetes pods, so that the pods can connect to pg at runtime. this is where it gets tricky... • is there some way to get

postgresql.ProviderArgs

out of the provider, during pulumi runtime? • is there some easy way to marshal/unmarshal

postgresql.ProviderArgs

, so that i can use a string stack export? keep in mind there are no pulumi types (input/output) defined for

postgresql.ProviderArgs

-- that's just a struct containing a bunch of pulumi types tia 🙏

Can someone link me an Golang example of how to create a CRD instance using Pulumi? I want to create an instance of Argos Application CRD…

I am currently using Pulumi to spin up a GKE cluster and deploy ArgoCD onto the cluster using the Argo Helm chart. As part of the setup I need to add a private GitRepo. This information is stored in a configmap (argocd-cm) which gets created by the Helm-chart… In order to add the private repo I would have to get the existing configmap (corev1.GetConfigMap( ….) ) and add the values to the data of the configmap. 1. is this really a good idea? 2. can I just use the GetConfigMap func and change the values…

HI , i try to create target groups .... and its doing it , pulumi return and list of obj "[]*lb.TargetGroup" in the struct i can see pulumi.StringOutput so my qustion is how do i get this data as string so i can use it in my next ... i need to commper the name of the TG to a tag in the LB and mach the TG to LB ... i want to do it automatic as i have list of TG i create and List of TB i create second Q all my resources are add suffix i try to disable it but is not working any one know how ?

Hi! I have a problem with pulumi cicd using github workflows and installing the google cloud provider. We use the google cloud and the kubernetes providers. The kubernetes provider automatically installs an appropriate version of the plugin, but the gcp provider is never installed. Can we make the gcp plugin version also install automatically? If not, what would be the recommended way of manually installing it? What we are doing is very close to what is described in this guide. This is the corresponding GitHub workflows yaml:

name: Pulumi

on:

- pull_request

jobs:

preview:

name: Preview

runs-on: ubuntu-latest

steps:

- name: checkout

uses: actions/checkout@v2

with:

fetch-depth: 1

- name: preview

uses: <docker://pulumi/actions>

with:

args: preview

env:

GOOGLE_CREDENTIALS: ${{ secrets.GOOGLE_CREDENTIALS }}

PULUMI_ACCESS_TOKEN: ${{ secrets.PULUMI_ACCESS_TOKEN }}

PULUMI_CI: pr

PULUMI_ROOT: infra/cluster/processing

This is the output from the preview step: 

Run <docker://pulumi/actions>

10 /usr/bin/docker run --name pulumiactions_99916e --label 3b3ac6 --workdir /github/workspace --rm -e GOOGLE_CREDENTIALS -e PULUMI_ACCESS_TOKEN -e PULUMI_CI -e PULUMI_ROOT -e INPUT_ARGS -e HOME -e GITHUB_JOB -e GITHUB_REF -e GITHUB_SHA -e GITHUB_REPOSITORY -e GITHUB_REPOSITORY_OWNER -e GITHUB_RUN_ID -e GITHUB_RUN_NUMBER -e GITHUB_ACTOR -e GITHUB_WORKFLOW -e GITHUB_HEAD_REF -e GITHUB_BASE_REF -e GITHUB_EVENT_NAME -e GITHUB_SERVER_URL -e GITHUB_API_URL -e GITHUB_GRAPHQL_URL -e GITHUB_WORKSPACE -e GITHUB_ACTION -e GITHUB_EVENT_PATH -e RUNNER_OS -e RUNNER_TOOL_CACHE -e RUNNER_TEMP -e RUNNER_WORKSPACE -e ACTIONS_RUNTIME_URL -e ACTIONS_RUNTIME_TOKEN -e ACTIONS_CACHE_URL -e GITHUB_ACTIONS=true -e CI=true -v "/var/run/docker.sock":"/var/run/docker.sock" -v "/home/runner/work/_temp/_github_home":"/github/home" -v "/home/runner/work/_temp/_github_workflow":"/github/workflow" -v "/home/runner/work/xxx/xxx":"/github/workspace" pulumi/actions preview

11 Logging in using access token from PULUMI_ACCESS_TOKEN

12 Logged in to <http://pulumi.com|pulumi.com> as xxx (<https://app.pulumi.com/> xxx)

13 Activated service account credentials for: [<mailto:pulumi-github-workflows-cicd@xxx.iam.gserviceaccount.com|pulumi-github-workflows-cicd@xxx.iam.gserviceaccount.com>]

14 Adding credentials for all GCR repositories.

15 WARNING: A long list of credential helpers may cause delays running 'docker build'. We recommend passing the registry name to configure only the registry you are using.

16 Docker configuration file updated.

`17 #### 🍹 `pulumi --non-interactive preview``

18 Previewing update (yyy/xxx):

19 [resource plugin kubernetes-2.3.0] installing

20

21 Downloading plugin: 0 B / 21.46 MiB    0.00%

22 Downloading plugin: 21.46 MiB / 21.46 MiB  100.00%

23 Downloading plugin: 21.46 MiB / 21.46 MiB  100.00%

24 Downloading plugin: 21.46 MiB / 21.46 MiB  100.00%

25 Downloading plugin: 21.46 MiB / 21.46 MiB  100.00% 0s

26 Moving plugin... done.

27

28 Permalink: <https://app.pulumi.com/xxx/previews/0472bcb1-a637-4708-96e9-21f0fe9862ff>

29 error: could not load plugin for gcp provider 'urn:pulumi:xxx::cadence::pulumi:providers:gcp::gcp-provider': no resource plugin 'gcp' found in the workspace or on your $PATH

Thanks in advance for any help.