typescript

then you want to look at pulumi-awsx and pulumi-cloud

oh, local?

i handled this by writing a bunch of code to let it select between live and localstack providers

no, i'm not aware of any pulumi-native stuff for local testing

pulumi up does not seem to be picking up my tsconfig.json paths. I also have ts-mocha setup and it picks up the paths. Is there something special I have to do?

@billowy-laptop-45963, do you write tests for your pulumi code? How is your experience, does it worth the effort?

@best-xylophone-83824 I don't actually use pulumi. I'm just playing around with it. Normally we use automated testing so I was curious what are the options for pulumi/typescript.

There’s a great blog post on our Pulumi blog about a good way to test

I can link

Life is nice with Pulumi! 😉 /cc @broad-dog-22463

Anybody else tackled roleMappings for EKS clusters and passing in dynamically retrieved objects such as roleARNs? Right now I’m passing them from an iam stack as outputs then retrieved for roleMappings via

.getOutput('roles').apply(etc etc)

. But part of me also wants the roleMappings to be a config block that passes into the cluster creation. How are most of y’all doing this?

How are pulumi security vulnerabilities announced?

Hello all. So I’m loving pulumi so far, but I can’t figure out the best way to deploy what should be fairly simple. I’m trying to deploy a few lambdas that make HTTP requests to a third party API. They are invoked directly by one of our microservices. There seems to be a few ways to do this, but the issue I was having is that I created an Object in TypeScript that had a few methods defined on it, and attempted to import it into the index.ts file. I was getting a lot of errors about serialization, etc. so I tried importing the code inside the closure of the function but still no luck. Does anyone know of a working example that demonstrates something like this? Thanks.

Quick note that we just posted an update to https://pulumi.statuspage.io/ about an increased error rate we're currently seeing impacting a small number of users. We are investigating.

why secret doesn't deserve it's own type? knowing which of pulumi.Output are secrets is hard sometimes

also resources would be able to declare some of their output as secrets by default and document it in APIdocs by using dedicated type for it, like https://www.pulumi.com/docs/reference/pkg/nodejs/pulumi/gcp/serviceaccount/#Key-privateKey

Did anybody start working on a gitlab-in-kubernetes pulumi module?

Blog post idea: provide an example how to run pulumi with short lived key for it's service account (https://cloud.google.com/iam/docs/creating-short-lived-service-account-credentials)

Hi, Recently I'm getting exceptions regarding empty resource ID. I can create resource but i can't recreate and destroy them. https://github.com/pulumi/pulumi/blob/master/pkg/resource/plugin/provider_plugin.go#L667 https://github.com/pulumi/pulumi/blob/master/pkg/resource/deploy/step.go#L248 I'm wondering if it can be caused by duplicated resource names. Anyone encounterd it? Thanks example message :

panic: fatal: An assertion has failed

goroutine 81 [running]:
<http://github.com/pulumi/pulumi/pkg/util/contract.failfast(...)|github.com/pulumi/pulumi/pkg/util/contract.failfast(...)>
        /home/travis/gopath/src/github.com/pulumi/pulumi/pkg/util/contract/failfast.go:23
<http://github.com/pulumi/pulumi/pkg/util/contract.Assert(...)|github.com/pulumi/pulumi/pkg/util/contract.Assert(...)>
        /home/travis/gopath/src/github.com/pulumi/pulumi/pkg/util/contract/assert.go:26
<http://github.com/pulumi/pulumi/pkg/resource/deploy.NewDeleteStep|github.com/pulumi/pulumi/pkg/resource/deploy.NewDeleteStep>(0xc001362180, 0xc0012a5800, 0xc0000c5300, 0x73)
        /home/travis/gopath/src/github.com/pulumi/pulumi/pkg/resource/deploy/step.go:237 +0x261
<http://github.com/pulumi/pulumi/pkg/resource/deploy.(*stepGenerator).GenerateDeletes(0xc0003d30e0|github.com/pulumi/pulumi/pkg/resource/deploy.(*stepGenerator).GenerateDeletes(0xc0003d30e0>, 0xc00007cc00, 0x2, 0x0)
        /home/travis/gopath/src/github.com/pulumi/pulumi/pkg/resource/deploy/step_generator.go:539 +0x884
<http://github.com/pulumi/pulumi/pkg/resource/deploy.(*planExecutor).Execute.func3(0xc001347980|github.com/pulumi/pulumi/pkg/resource/deploy.(*planExecutor).Execute.func3(0xc001347980>, 0xc0009dbba0, 0xc000c2e290, 0x18bfee0, 0xc000c32040, 0x18bfee0, 0xc000c32000, 0x10000, 0x1, 0xc0004e9c20)
        /home/travis/gopath/src/github.com/pulumi/pulumi/pkg/resource/deploy/plan_executor.go:155 +0x3d5
<http://github.com/pulumi/pulumi/pkg/resource/deploy.(*planExecutor).Execute(0xc0009dbba0|github.com/pulumi/pulumi/pkg/resource/deploy.(*planExecutor).Execute(0xc0009dbba0>, 0x18bfee0, 0xc000c32000, 0x18c01e0, 0xc000391810, 0x7fffffff, 0x10000, 0xc0007f5f01, 0x0, 0x0)
        /home/travis/gopath/src/github.com/pulumi/pulumi/pkg/resource/deploy/plan_executor.go:192 +0x511
<http://github.com/pulumi/pulumi/pkg/resource/deploy.(*Plan).Execute(...)|github.com/pulumi/pulumi/pkg/resource/deploy.(*Plan).Execute(...)>
        /home/travis/gopath/src/github.com/pulumi/pulumi/pkg/resource/deploy/plan.go:330
<http://github.com/pulumi/pulumi/pkg/engine.(*planResult).Walk.func1|github.com/pulumi/pulumi/pkg/engine.(*planResult).Walk.func1>(0x18c01e0, 0xc000391810, 0xc000391760, 0x18bfee0, 0xc000c32000, 0x14daf01, 0xc000c2e280, 0xc0013477a0)
        /home/travis/gopath/src/github.com/pulumi/pulumi/pkg/engine/plan.go:195 +0x13b
created by <http://github.com/pulumi/pulumi/pkg/engine.(*planResult).Walk|github.com/pulumi/pulumi/pkg/engine.(*planResult).Walk>
        /home/travis/gopath/src/github.com/pulumi/pulumi/pkg/engine/plan.go:186 +0x11b

Hello, I am trying to import a cloudsql postgres instance on gcp. I am getting

warning: inputs to import do not match the existing resource; importing this resource will fail

but the details view doesn't highlight any difference. Does anyone know how to get a more verbose output and track down what my input is missing?

Our Pulumi projects are growing at a fast clip. Thanks to the fact that it’s Just Code we can use abstractions and keep everything organized, but if we stay on our current pace the total number of resources is likely to reach the thousands, eventually. Does anyone have experience with stacks that big? Is this something we need to think about or can Pulumi handle this just fine?

we started to split stuff into 'layers' so that e.g. deploying apps to kube didnt involve checking or refreshing any of the infrastructure around it

also less risk of accidentally nuking everything

Learning about ComponentResource, I checked https://github.com/pulumi/pulumi-eks/blob/master/nodejs/eks/cluster.ts as an example. Class have multiple public properties, but only one of them is

pulumi.Output<>

and it is the only passed to

this.registerOutputs()

, why is that?

how can I configure ambient provider from index.ts itself without reparenting all resources under ComponentResource? I have bootstrap pulumie project with a single stack which creates GCP projects and service accounts with key made available in stack output. Idea is that individual pulumi projects should then reference that stack get key from it's output and then configure ambient provider to use it

I can probably do that with bash, but would prefer to do it natively

checks time in Seattle

0704 😉