Hi, is there a good way to get the AWS provider used by an AWS resource?

i noticed that all dynamic resources are created with Type =

pulumi-nodejs:dynamic:Resource

- Is there some way to override this? It makes it hard to tell what’s happening in previews & means i have to do extra work to namespace resources

This message was deleted.

i’m using

cloud.timer.interval

& logs show up in cloudwatch, but aren’t showing up in

pulumi logs -f

(other logs are) - how does pulumi logs know what to monitor? and is this a bug in the cloud library maybe?

Hi, just curious how the Pulumi developers/users think about security of e.g. downstream dependencies of Pulumi? Is it a 'downstream authors are trusted' kind of thing, or does some type of vetting on downstream dependencies happen? I am wondering this because there are lots of downstream deps and Pulumi working directly with infrastructure opens new attack vectors (secret access, etc). Thoughts? http://npm.broofa.com/?q=@pulumi/pulumi

Is it possible to make use of "runtime" functionality locally, or will it only work in a lambda? My use case is, I need to run an ecs task as part of a test - it would be convenient if I could just use the task.run method on the resource instead of having to use aws' sdk

Ah Joe's recent blog post may have answered my question, looks like it's something available in golang but not yet implemented for other SDK's

I can’t find any examples of using host-based routing with ALBs in pulumi, specifically with the awsx module. Has anyone out there done this successfully?

Hi, I’m following https://github.com/pulumi/pulumi-tf-provider-boilerplate for datadog provider and when running

make ensure

getting

ENSURE:
GO111MODULE=on go mod vendor
/bin/bash: gomod-doccopy: command not found
make: *** [ensure] Error 127

I have a typescript pulumi stack exporting some stack outputs — is there a way I can use the exports from another (non-pulumi) typescript script in a type safe way? Will it work to just import the module in which the stack is defined?

Hi, Here’s a DataDog provider https://github.com/lemonade-hq/pulumi-datadog . Should we transfer the ownership?

Tags don't seem to be available when using

eks.Cluster

. I have a need to add tags to the cluster and all related/dependent components, but there's not tags attribute available. Is there any workaround for this?

39 /*
 40  * EKS cluster creation
 41  */
 42 export const cluster = new eks.Cluster(config.environment + "-eks-cluster", {
 43     version: config.kubernetesVersion,
 44     vpcId: config.vpcId,
 45     subnetIds: config.subnetIds,
 46     instanceType: config.instanceType,
 47     tags: {
 48       Project: "Mustang",
 49     },
 50     nodeAmiId: config.amiId,
 51     nodePublicKey: config.workerNodePublicKey,
 52     nodeRootVolumeSize: 200,
 53     desiredCapacity: config.desiredNodes,
 54     maxSize: config.maxNumberOfNodes,
 55     minSize: config.minNumberOfNodes,
 56     nodeUserData: userData,
 57     deployDashboard: false,
 58     vpcCniOptions: {



$ pulumi up
Previewing update (redacted):

     Type                 Name                                              Plan     Info
     pulumi:pulumi:Stack  eks-cluster-online-sandbox-vpc           1 error

Diagnostics:
  pulumi:pulumi:Stack (eks-cluster-online-sandbox-vpc):
    error: Running program '/home/tsi/eshamay/git/cs/infrastructure/pulumi/eks-cluster' failed with an unhandled exception:
    TSError: ⨯ Unable to compile TypeScript:
    cluster.ts(47,5): error TS2345: Argument of type '{ version: string; vpcId: string; subnetIds: Output<UnwrappedArray<string>>; instanceType: InstanceType; tags: { Project: string; }; nodeAmiId: string; nodePublicKey: string; nodeRootVolumeSize: number; ... 6 more ...; roleMappings: { ...; }[]; }' is not assignable to parameter of type 'ClusterOptions'.
      Object literal may only specify known properties, and 'tags' does not exist in type 'ClusterOptions'.

@white-balloon-205 In https://github.com/pulumi/pulumi-eks/blob/master/nodejs/eks/cluster.ts#L92 - both

instanceRole

and

instanceRoles

are outputs and thus the expression fails even though none of them is set

Is it possible to use

link:../whatever

in a pulumi package.json for a yarn project? I get

Could not include required dependency

when i try.

I'm still getting a "Failed to refresh the Token for request" when building my Azure infrastructure. It seems to happen randomly as its completely inconsistent even when working from the same Pulumi code/stack. Has anyone else really never seen this?

It has happened once out of the five builds I've done today

In the last 1-3 days

pulumi up

became very slow. From less than 30-50 seconds each run is now taking ~3 minutes. Is it a known issue?

Not sure if this has been reported already but I'm seeing SSL issues on https://www.pulumi.com/

NET::ERR_CERT_COMMON_NAME_INVALID

has anyone got auth working with the

@pulumi/newrelic

package? I'm hitting errors like this

error: Plan apply failed: The API key provided is invalid

.

We are trying to upgrade our GKE cluster and the operations keeps timing out at the pulumi end with

10

mins timeout. We are stuck and not able to proceed.

What are the options available for this?

error: could not load plugin for azure provider 'urn:pulumi:subscription::subscription::pulumi:providers:azure::default': no resource plugin 'azure-v0.17.3' found in the workspace or on your $PATH, install the plugin using `pulumi plugin install resource azure v0.17.3`

i'm getting that for a

pulumi refresh

on CI - never had it do this before with package updates. Any solution?

I’m following the example here, but this line doesn’t work, is there any way I can use the value in the output? https://github.com/pulumi/pulumi-aws/blob/master/sdk/nodejs/ec2/getSubnetIds.ts#L24

Hello. Has anyone ever created a Private GKE Cluster? I'm looking for the the

privateCluster

option (as described in https://pulumi.io/reference/pkg/nodejs/@pulumi/gcp/container/#Cluster-privateCluster) but I can't find it in the

ClusterArgs

(https://github.com/pulumi/pulumi-gcp/blob/master/sdk/nodejs/container/cluster.ts), despite seeing it in the

GetClusterResult

interface (https://github.com/pulumi/pulumi-gcp/blob/master/sdk/nodejs/container/getCluster.ts#L105). If I set only the

privateClusterOptions

I get the following error:

gcp:container:Cluster (my-cluster):
    error: Plan apply failed: googleapi: Error 400: Private cluster must be enabled if private endpoint is enabled., badRequest

Anyone ever had this? Am I missing something? Thanks in advance.

what’s the recommended way to inject secrets in to lambda functions produced by pulumi / pulumi-cloud ? aws lambdas can have environment variables encrypted at rest, which seems like i’d want to take advantage to that, but open to alternatives. is there an example of this somewhere?

Having a minor UI/usability issue with https://app.pulumi.com/MINDBODY-Platform

Depending on your git remote your stack info will show up in 1 of 3 places:

if https://orgname.visualstudio.com vs https://dev.azure.com/orgname/repo/_git/program vs ssh://dev.azure.com/orgname/repo/program

Hello all, I’m running into a weird issue with ACM Cert validation failing to complete in time. It seems to be because, The DNS records aren’t being setup in Route 53. I had to go into the aws console, and click the button to set them up. So somehow

new aws.acm.CertificateValidation()

is not doing this by default?

I noticed that the homebrew formula for pulumi is still at `v0.17.6`: https://github.com/Homebrew/homebrew-core/blob/963a3c49ce6050b1ee6a891bfe50489f4af59005/Formula/pulumi.rb#L5 Is there an ETA for updating the formula for `v0.17.7`/`v0.17.8`?