I am trying to unwrap aws.rds.Instance type to use it as an input in a dynamic provider, but Unwrap<aws.rds.Instance> produces aws.rds.Instance (nothing is unwrapped), what I am doing wrong?

also this simple example from the pulumi programming model produces a trace: https://www.pulumi.com/docs/intro/concepts/programming-model/#example-random

Newbie here , exploring Pulumi . Is Pulumi only for bootstrapping Cloud infra (and /or subsequent tear down) or can it also be dropped into existing Cloudscape , and be able to do wholistic inventory of resources, accounts , costs etc ? All the examples and webinars seem to concentrate on bootstrapping , so if possible any link to examples or docs would be great .

Is it possible to tell the error/ret code due to

--expect-no-changes

from other errors?

Hi there, More of a design question this time. We have an API Gateway in our infrastructure, all deployed with pulumi (there’s a chart for this gateway), and to set it up, we need to run a bunch of REST API calls against it (that’s a one shot init), so with a script. I can obviously do it manually with a port-forward, but the ultimate goal would be to do it with pulumi. How would one do that the pulumi way ? We use TS so we can do the script in TS with pulumi, but what would be the right way to use it ? I was thinking a

CustomResource

to run it once at the create state and never again, but we never did one, we always do

ComponentResource

. Does anyone would have examples or some thoughts about that please ?

Hi guys, I was tring to use 

pulumi refresh

 to sync k8s cluster's status to local stack. But when I tried to apply the previous code by using 

pulumi up

, it tells me there is no updates. Is there anything I misunderstand about 

pulumi refresh

?

this might be a dumb question, but if I am building a ComponentResource should I take inputs to my class Input<T> or T? For example I can pass a string into a resource that expects input<string>, so what is the benefit of using Input?

hi there, excellent product, just started playing with it, already loving it 🙂 the question I have for you guys is... is it possible to set API Gateway as Cloudfront origin? I tried the following, and the error I'm receiving can be seen in the console (see sshot please) kinda confused what would be the way to achieve, so any help would be appreciated 🙂

@echoing-match-29901 The pulumi website seems to be broke.

Docs and blog sub-domains seem to be working

@ancient-megabyte-79588 i'm not seeing that - what response do you get when you try download that js file with

cURL

for example? do you see any response code or error log in the chrome console that might help us debug?

This may be an internal network problem on our side. I'm on the VPN right now. I haven't had this problem before though. 😞

My g/f beside me (not on VPN) is able to get to Pulumi.com and it renders ok

My phone also gets there fine. Sorry for the alarm. I'll reach out to internal infrastructure team.

Most of the site content does download though.

Just not those three files (css bundle and 2 js bundles)

Everytime I run

pulumi preview

it wants to update my digitaloceans k8s cluster. I have

version

set to latest but I've tried with

1.18

as well, same results.

I am using the Go Integration Testing library as described here: https://www.pulumi.com/docs/guides/testing/integration/#full-examples I get the following error:

=== RUN   TestPulumiInt
[ Pulumitest ] Launching test...
[ Pulumitest ] Launching Pulumi ProgramTest...
=== PAUSE TestPulumiInt
=== CONT  TestPulumiInt
    TestPulumiInt: program.go:546: Skipping: PULUMI_ACCESS_TOKEN is not set
--- SKIP: TestPulumiInt (0.00s)

Is there a way to configure the local backend when running integration tests? From the Code I see that only CloudUrl is supported: https://github.com/pulumi/pulumi/blob/b77ec919d4957cae450003bd8d28cf634a9703fe/pkg/testing/integration/program.go#L1077

Hi, I have 2 roles, which is basically the un-authenticated and authenticated roles for Identity Pool in AWS, there are policies attached to it, when I run pulumi destroy I get the following error

Folks I want to share with you something extremely precious I have found out that can help a lot of people trying to use ephemeral project on GCP : https://github.com/terraform-google-modules/terraform-google-project-factory . Especially IAM wise, you will face a lot of challenges if you are not GCP Gurus. I am doing the same approach with Pulumi and it is working great

Hi, I want to know how I can create a new resource based on a previously created resource, without hitting the Output vs String problem. For example:

from pulumi_azure import core
resource_group = core.ResourceGroup("foo")
core.ResourceGroup(resource_group.name + "bar")

With this I get the error

unsupported operand type(s) for +: 'Output' and 'str'

, which is expected. And now:

from pulumi_azure import core
resource_group = core.ResourceGroup("foo")
core.ResourceGroup(Output.concat(resource_group.name, "bar"))

But with this I get the error

TypeError: Expected resource name to be a string

. Same thing happens if I use an

apply

. So is there any way to do it? What is the alternative? Thank you.

Dear all, I have a conceptual question: Do I understand correctly, that pulumi only updates the state based on the code that is executed, but does not update the state based on changes that might be applied manually e.g. on AWS using the AWS UI? If that is the case, how does one assure that those manual changes are not accidentally overwritten?