thanks, I'll take a look!

I want to create a custom component provider, starting from the Go template here https://github.com/pulumi/pulumi-component-provider-go-boilerplate I cloned this repo and ran

make install_nodejs_sdk

on my Mac, but even without a single change, this build command fails:

$ make install_nodejs_sdk
rm -rf sdk/nodejs
cd provider/cmd/pulumi-gen-xyz && go run . nodejs ../../../sdk/nodejs /Users/r.de.smet/Projects/pulumi-component-provider-go-boilerplate/schema.yaml
cd sdk/nodejs/ && \
                yarn install && \
                yarn run tsc --version && \
                yarn run tsc && \
                cp ../../README.md ../../LICENSE package.json yarn.lock ./bin/ && \
                sed -i.bak -e "s/\${VERSION}/0.0.1/g" ./bin/package.json && \
                rm ./bin/package.json.bak
yarn install v1.22.17
warning package.json: No license field
info No lockfile found.
warning @pulumi/xyz@${VERSION}: No license field
[1/4] 🔍  Resolving packages...
warning @pulumi/aws > read-package-tree@5.3.1: The functionality that this package provided is now in @npmcli/arborist
warning @pulumi/aws > @pulumi/pulumi > read-package-tree@5.3.1: The functionality that this package provided is now in @npmcli/arborist
warning @pulumi/aws > aws-sdk > querystring@0.2.0: The querystring API is considered Legacy. new code should use the URLSearchParams API instead.
warning @pulumi/aws > aws-sdk > url > querystring@0.2.0: The querystring API is considered Legacy. new code should use the URLSearchParams API instead.
warning @pulumi/aws > aws-sdk > uuid@3.3.2: Please upgrade  to version 7 or higher.  Older versions may use Math.random() in certain circumstances, which is known to be problematic.  See <https://v8.dev/blog/math-random> for details.
[2/4] 🚚  Fetching packages...
[3/4] 🔗  Linking dependencies...
[4/4] 🔨  Building fresh packages...
success Saved lockfile.
✨  Done in 28.81s.
yarn run v1.22.17
warning package.json: No license field
$ /Users/r.de.smet/Projects/pulumi-component-provider-go-boilerplate/sdk/nodejs/node_modules/.bin/tsc --version
Version 3.9.10
✨  Done in 0.35s.
yarn run v1.22.17
warning package.json: No license field
$ /Users/r.de.smet/Projects/pulumi-component-provider-go-boilerplate/sdk/nodejs/node_modules/.bin/tsc
✨  Done in 12.38s.
yarn unlink xyz
yarn unlink v1.22.17
error No registered package found called "xyz".
info Visit <https://yarnpkg.com/en/docs/cli/unlink> for documentation about this command.
make: *** [install_nodejs_sdk] Error 1
$

What am I missing here? Also, if I want to change the setup to have my own package name, generated in my own namespace (npm), what files should I update manually and which ones are generated? Not much documented about all of this.

Hi, is there a standard way to convert

StringPtrOutput

to

StringOutput

(and default to zero value?) Right now we wrote our own helper func, but it looks strange

StringPtrOutput

has a

Elem()

method that does this ^

Hi, I am deploying virtual machine using go automation api . While testing and deploying machine there was panic and rerunning the program I am Now getting following error when I try to either delete stack or create new machine

error: could not create stack: the stack is currently locked by 1 lock(s). Either wait for the other process(es) to end or manually delete the lock file(s).

the only Way I found was to go and delete the .json lock file manually from the stack folder . Just wondering how I can manage and check if the stack is lock using Automation api and unlock the locked stack using automation api , instead of the deleting the stack file manually .

i just updated my ws with a

go mod tidy

but now i'm getting:

pulumi:providers:aws (default):

`error: no resource plugin 'aws-v4.37.4' found in the workspace or on your $PATH, install the plugin using `pulumi plugin install resource aws v4.37.4`` but that doesn't work:

$ pulumi plugin install resource aws v4.37.4

[resource plugin aws-4.37.4] installing

error: [resource plugin aws-4.37.4] downloading from : 403 HTTP error fetching plugin from <https://get.pulumi.com/releases/plugins/pulumi-resource-aws-v4.37.4-linux-amd64.tar.gz>

any idea what's going on?

Hello - when I call

pulumi.All(a, b, c).ApplyT(myFunc)

and

c

is a pulumi secret, the resulting output of

ApplyT

is also a secret, right? This doc talks about

ApplyT()

but not

All()

Hello folks, hoping to get some guidance from people here that have experience using the pulumi golang sdk and preferably also with the typescript sdk. After having used the typescript sdk for aws and eks for some time - it has been quite a seamless experience, and things work "as advertised" for the most part. I have also made some small contributions which were graciously accepted into the upstream code. Now I am trying to use the golang sdk (mainly so that we can get wider adoption in house), and the experience seems quite degraded compared to ts. Also, it seems like the golang libs still rely on typescript-written plugins (e.g. aws, kubernetes) and so its not as compelling to use golang. Additionally, (eks) stacks created with golang seem to be more "high level" - as in - they are not aware of lower-level resources in the stack. This could be partially due my being relatively new to Golang itself. Any well-informed opinions on this? Are developers finding the Golang sdk to be mature and usable/maintainable enough to justify using it over TS for prod infra? Thanks in advance.

I don't have experience with the TS one, but we use Go extensively and find it mature. There are Go specific pain points, mostly around Go’s restrictive type system. I think for someone coming from TS specifically it can be painful. I'm not sure what you mean about plug-ins written in TS in reference to above, or access to the lower level resources.

Hi @bored-table-20691 thanks for the response! Here is what I mean. This is a "pure" Go stack, starting with no pulumi plugins installed. When I

up

the stack, it downloads the plugins it needs.. which.. are in typescript, not Go / native:

➜  eks_go git:(feature/eks_go_aws) ✗ pulumi plugin ls
NAME  KIND  VERSION  SIZE  INSTALLED  LAST USED

TOTAL plugin cache size: 0 B
➜  eks_go git:(feature/eks_go_aws) ✗ ls ~/.pulumi/plugins/
➜  eks_go git:(feature/eks_go_aws) ✗
➜  eks_go git:(feature/eks_go_aws) ✗ ls
eksCluster.go  go.mod  go.sum  main.go  nodeGroups.go  Pulumi.eks-test.yaml  Pulumi.yaml
➜  eks_go git:(feature/eks_go_aws) ✗ pulumi up
Please choose a stack, or create a new one: eks-test
Previewing update (eks-test):
[resource plugin kubernetes-3.16.0] installing
Downloading plugin: 0 B / 30.27 MiB [----------------------------------]   0.00%[resource plugin eks-0.37.1] installing
Downloading plugin: 656.37 KiB / 656.37 KiB [=======================] 100.00% 0s
[resource plugin aws-4.37.4] installing
Downloading plugin: 30.27 MiB / 30.27 MiB [=========================] 100.00% 1s
Downloading plugin: 96.60 MiB / 96.60 MiB [=========================] 100.00% 2s
     Type                     Name              Plan       Info
 +   pulumi:pulumi:Stack      eks_go-eks-test   create     3 errors; 9 messages
 +   └─ pulumi:providers:aws  test_east1  create

Diagnostics:
  pulumi:pulumi:Stack (eks_go-eks-test):
    error: Error: invocation of pulumi:pulumi:getResource returned an error: unknown resource urn:pulumi:eks-test::eks_go::aws:iam/role:Role::eksEc2Role
        at Object.callback (/home/vikas/.pulumi/plugins/resource-eks-v0.37.1/node_modules/@pulumi/runtime/resource.ts:139:40)
        at Object.onReceiveStatus (/home/vikas/.pulumi/plugins/resource-eks-v0.37.1/node_modules/@grpc/grpc-js/src/client.ts:338:26)
        at Object.onReceiveStatus (/home/vikas/.pulumi/plugins/resource-eks-v0.37.1/node_modules/@grpc/grpc-js/src/client-interceptors.ts:426:34)
        at Object.onReceiveStatus (/home/vikas/.pulumi/plugins/resource-eks-v0.37.1/node_modules/@grpc/grpc-js/src/client-interceptors.ts:389:48)
        at /home/vikas/.pulumi/plugins/resource-eks-v0.37.1/node_modules/@grpc/grpc-js/src/call-stream.ts:276:24
        at processTicksAndRejections (node:internal/process/task_queues:78:11)

I don’t think my Go projects match what you’re seeing.

What does your

Pulumi.yaml

in that directory look like?

(also we shoudl thread responses, apologies)

Hi Team , I am deploying EC2 machine using golang api. After provisioning we would like to confirm if the machine is fully provisioned before deploying other services. Is there a way to check using pulumi automation api to confirm that machine is fully provisioned?

What's the conventional way to bubble up errors from within an

ApplyT

? I noticed that the function can return an error...but it doesn't seem to have any effect. Are we supposed to call

panic(err)

within the interpolation function? And if so, is there a risk of that

panic

causing issues with the pulumi state or the infra state?

Any article on how to make sure you don't get nil pointer errors with pulumi and Go, like a static check or something that can help flag when I forget to make sure something was a pointer?

Team, trying to setup some

wafv2

objects on AWS. This `pulumiTypes.go` file is unusually big at 70MB and 7 million lines. My local

go build

on Mac takes 60 secs. Pulumi preview mostly hangs, sometimes works. All Github Action containers crash because of their provisioned CPU/mem. It is just

wafv2

that is giving problems. Provisioning other objects are mostly fine both locally and on CD pipelines. Any reason on why this is such an outlier and ways to work around this?

is there a way of accessing component children from the test? We're trying to register all of them via MockMonitor, but don't seem to find a way to access it later

Is there a way to add arbitrary HTTP headers to a provider? Context: I'm using the openstack provider and need to go through a proxy that requires additional auth via extra headers.

Hi guys, can someone point out what I'm missing here? I'm trying to import a VpcId from a VPC object into a subnet declaration with the AWS provider.

vpc, err := ec2.NewVpc(ctx, "partseeker-dev", &ec2.VpcArgs{
			CidrBlock: pulumi.String("10.0.0.0/16"),
		})
		assertNoErr(err)

		publicA, err := ec2.NewSubnet(ctx, "publicA", &ec2.SubnetArgs{
			VpcId:            pulumi.String(vpc.ID()),    //<---- here
			AvailabilityZone: pulumi.String(appAZ),
			CidrBlock:        pulumi.String("10.0.10.0/24"),
		})

VpcId needs a string, and I thought vpc.ID() returns a string... but it returns the type IDOutput. What should I be doing instead?

Is there a better way to write json iam policies than using a bunch of []map[string]interface{}{ everywhere and pull it together with json.Marshal?

Hello Community !! Posting the same question I asked on #kubernetes https://pulumi-community.slack.com/archives/CRFURDVQB/p1647278981882749

How are people structuring there pulumi Go project folders? What directories do you use? Go recommends folders like CMD, Internal.

Hi all. I'm using the Go SDK to create a small util to set up a stack using a local backend. For example:

project := auto.Project(workspace.Project{                                                                                                                                                                
        Name:    tokens.PackageName(projectName),                                                                                                                                                         
        Runtime: workspace.NewProjectRuntimeInfo("go", nil),                                                                                                                                              
        Backend: &workspace.ProjectBackend{                                                                                                                                                               
                URL: "file://~/.pulumi-local",                                                                                                                                                               
        },                                                                                                                                                                                                })                                                                                                                                                                                                        
                                                                                                     
s, err := auto.UpsertStackInlineSource(ctx, stackName, projectName, deployFunc(configPath), project, secretsProvider, stackSettings, envvars)

My issue is that if

~/.pulumi-local

directory doesn't exist, this will fail with

stderr: error: unable to open bucket file:///home/tim/.pulumi-local: stat /home/tim/.pulumi-local: no such file or directory

. If the directory exists, the code works and the stack is updated without issue. Looking through the pulumi CLI code, the 'login' function is what would normally create the local backend directory if it doesn't exist, but I can't seem to find this functionality exposed by the SDK.

Hi I am deploying EKS cluster using pulumi . I see there are different Go SDK I can use to deploy cluster .

<http://github.com/pulumi/pulumi-aws/sdk/v4/go/aws/eks|github.com/pulumi/pulumi-aws/sdk/v4/go/aws/eks>

or

"<http://github.com/pulumi/pulumi-eks/sdk/go/eks|github.com/pulumi/pulumi-eks/sdk/go/eks>"

Just wondering which SDK should i use for EKS cluster provisioning ?

Perhaps I need to ask the question in here: I'm trying to get the Instance Role from a newly created EKS cluster (i.e. created in the same block of code. But I'm struggling to dereference the Output from the cluster object to give me the name of the Instance Role so I can add some permissions to it. https://pulumi-community.slack.com/archives/C84L4E3N1/p1649445617086939?thread_ts=1649258358.357719&cid=C84L4E3N1 Describes where I've got to. I'd be very grateful for any insight as to how I can resolve this.

Hi. Can someone give me an example of how to use https://www.pulumi.com/registry/packages/eks/api-docs/managednodegroup/#nodegrouptaint ?

Hi, how can I serialize multiple iam.NewUser creation with iam.NewGroupMembership? The following code works most of the time but I’d like to implement a better serialization between these two resources:

func createUsers(ctx *pulumi.Context,
	group mcfg.Group,
	globalTags map[string]string) ([]pulumi.StringOutput, error) {

	// the order of users in the list is important to avoid
	// unnecessary updates of GroupMembership and to serialize its update
	// with NewUsers exectued in a loop.
	sort.Strings(group.Users)

	var newUsersNames []pulumi.StringOutput
	var newUser *iam.User
	var err error
	for _, user := range group.Users {

		newUser, err = iam.NewUser(ctx, user, &iam.UserArgs{
			Name: pulumi.String(user),
			Path: pulumi.String("/perf/"),
			Tags: pulumi.ToStringMap(globalTags),
		})
		if err != nil {
			return nil, err
		}

		newUsersNames = append(newUsersNames, newUser.Name)
	}

	if len(newUsersNames) > 0 {

		ngMemberName := fmt.Sprintf("%s-%s", group.Name, "mozart")

		// a hack to serialize NewUsers with NewGroupMembership
		_, err := iam.NewGroupMembership(ctx, ngMemberName, &iam.GroupMembershipArgs{
			Users: pulumi.ToStringArray(group.Users),
			Group: pulumi.String(group.Name),
		}, pulumi.DependsOn([]pulumi.Resource{newUser}))

		if err != nil {
			return nil, err
		}
	}

	return newUsersNames, nil
}

My initial attempt was to convert

newUsersNames - []pulumi.StringOutput

to

[]string

using ApplyT on each of the list elements. It works but then the creation of

NewUsers

and

NewGroupMembership

is not serialized. In the above version I just pass the list of users which is provided by a function argument to

GroupMembershipArgs

and try to create a dependency on the last

NewUser

resource created in the loop. Is there a better way of doing this?

Hi I am using the Automation to create the aws EKS cluster.The pulumi code is expose as the rest interface . if we send the single request to create cluster it works fine . However when we send multiple requests to create multiple clusters on the same stack , it produces an error saying that stack is locked and cannot proceed further until the one task is complete . Is there a way where we can send multiple request in Parallel. We are using go code for performing this task

I found an unexpected behavior yesterday while porting some Pulumi code over to a set of new AWS accounts. As part of porting the code over to the new accounts, I switched from using

AWS_PROFILE

in the environment (set via

.envrc

and

direnv

) to specifying the AWS profile in a Pulumi configuration value (i.e.,

pulumi config set aws:profile <name>

). This worked for everything except a Pulumi program that sets up VPC peering between VPCs in two different regions. The error it returned was related to the AWS provider that operates in the “destination” region (one provider operates in the “source” region to set up peering-related objects, and a second provider operates in the “destination” region to set up the corresponding objects on the other side of the connection). Is this expected behavior? I haven’t dug deep into fixing it yet, but I’m guessing there’s something about the way I’m setting up the destination provider that isn’t working with/respecting the

aws:profile

configuration value. (All this is written in Go, hence why I’m posting here.)