they are still unaware https://www.githubstatus.com

ok, everyone is down 😉

Don't worry, it's not just pulumi

status page mentions it is back and it works again for me

yea, US-West died today... whatever they're rolling out will probably hit EU next. Can't wait!

Hi, I'm trying to setup a new trigger on a lambda function in pulumi. I want the source to be a cloudwatch log stream. Is this possible with pulumi? Also, if this is not the right channel please lmk and I'll move the quest there 🙂

Is anyone else trying to use the pulumi-kafka package to manage topics and acls in confluent cloud? I am attempting to create topics with a provider that looks like this:

provider = kafka.Provider(
                "kafka-provider",
                bootstrap_servers=confluent_environment.apply(
                    lambda e: [e.bootstrap_servers.lstrip("SASL_SSL://")]
                ),
                opts=opts,
                sasl_mechanism="plain",
                sasl_username=confluent_environment.apply(
                    lambda e: e.environment_credentials.api_key
                ),
                sasl_password=confluent_environment.apply(
                    lambda e: e.environment_credentials.api_secret
                ),
                tls_enabled=True,
                timeout=120,
            )

and getting this error:

kafka:index:Topic (logs-topic):
    error: 1 error occurred:
        * kafka server: The client is not authorized to access this topic.

I have given the service account attached to my API key the

CloudClusterAdmin

entitlement and have confirmed with the confluent CLI that the API key can create a topic in the cluster... Any ideas what might be going wrong here?

Hi all - quick question I'm trying to import some Azure Native resources by ID, specifically a VNet right now. The Subnets are throwing up some mismatches which are mostly erroneous (saying it needs to update e.g. RouteTable but the IDs are identical). My main query is though how do I ignore properties that sit below an array? I can use

ignoreChanges

to set individual elements e.g.

"subnets[0].routeTable"

but I want to do it for all with some syntax like

"subnets[*].routeTable"

. I've tried a few combos like [] and [*] or no brackets at all, but no luck so far. What have I missed?

Hey Guys does anyone know how to get a vault secret using azure-native

Hello! I wrote a blog on incrementally migrating from Terraform/Terragrunt to Pulumi if anyone is interested 😄: https://news.ycombinator.com/item?id=29582542 To do so, we created a Terraform Provider for the Pulumi cloud, which is open sourced here: https://github.com/transcend-io/terraform-provider-pulumi/

Hey guys! Does anyone know in regards to using pulumi aliases, once you configure them for a resource and deploy, can you then remove the alias configuration from the code? So does the alias field need to be set for eternity still, after having deployed once with it? https://www.pulumi.com/docs/intro/concepts/resources/#aliases https://www.pulumi.com/blog/cumundi-guest-post/

I'm running into some trouble with the AWS-Native package in golang, I need an AWS Role so I copied the example here: https://www.pulumi.com/registry/packages/aws-native/api-docs/iam/role/#example-2 thusly

eksRole, err := iam.NewRole(ctx, "eksRole", &iam.RoleArgs{
                        AssumeRolePolicyDocument: pulumi.Any{
                                Version: "2012-10-17",
                                Statement: []map[string]interface{}{
                                        map[string]interface{}{
                                                "effect": "Allow",
                                                "principal": map[string]interface{}{
                                                        "service": []string{
                                                                "<http://eks.amazonaws.com|eks.amazonaws.com>",
                                                        },
                                                },
                                                "action": []string{
                                                        "sts:AssumeRole",
                                                },
                                        },
                                },
                        },
                        Path: pulumi.String("/"),
                        ManagedPolicyArns: pulumi.StringArray{
                                pulumi.String("arn:aws:iam::aws:policy/AmazonEKSClusterPolicy"),
                                pulumi.String("arn:aws:iam::aws:policy/AmazonEKSServicePolicy"),
                        },
                })
                if err != nil {
                        return err
                }

But when I try and run that I get the following errors.

Diagnostics:
  pulumi:pulumi:Stack (EKS-EKS-test):
    # EKS
    ./main.go:77:30: pulumi.Any is not a type
    ./main.go:77:30: invalid composite literal type func(interface {}) pulumi.AnyOutput
 
    error: an unhandled error occurred: program exited with non-zero exit code: 2

I have spent some time checking I copied the example well enough and I can't find any syntax errors, also when I copy the code from the example verbatim into my main.go and try to run it I get similar errors. Did something change about pulumi.Any? Is there an error in the example? Please help!

Hi all, is it possible to have a Pulumi project to allocate resources be broken down into multiple files and folders? similar to how one would structure a Terraform project... For example, inside

dev

I have the file

index.ts

and then inside

dev

I also have another folder

iam

and inside it I have another

index.ts

file. I need

pulumi up

to execute both files. Is it possible?

Hi all, is Pulumi based on Terraform?

Is there a way to get the org name pre-filled for the SSO login? I mean something like

<https://app.pulumi.com/signin/sso?reason=401&orgName=><myorgname>

Hello All, I have a project that I have been deploying in Python3.6 for some time. I am deploying a container to aws ecs. I installed the latest version of Pulumi, v3.19.0 I created a virtual environment in Python 3.8 pulumi==3.20.0 pulumi-aws==4.33.0 pulumi-docker==3.1.0 When I run pulumi up, I get the message error: failed to discover plugin requirements: exit status 1 So, same code works in Python3.6 does not work in Python3.8 and the error message is vague. Here is the code I am trying to run

from pulumi import export
import pulumi_aws as aws


# Create an Elastic Container Service (ECS) cluster in the default VPC

cluster = aws.ecs.Cluster(
    "app-cluster",
    settings=[aws.ecs.ClusterSettingArgs(name="containerInsights", value="enabled")],
)

Also does not work in Python 3.9
Also tried all the latest pip install libraries with the same message
Also tried a brand new project, still same message

hello, can pulumi revert manual changes? I tested with k8s deployment, when manually changing replicas count, 'pulumi up' don't rollback this update. Is this expected behavior?

anyone can help me... im trying to integrate gitlab cicd with pulumi and when im runing the #pulumi up comand i get a [Failed to parse python version: ''] at console... but dont know how to debug it

Hello everyone 👋 Our team using pulumi on Azure, and I would like to use the

pulumi import

cli command to import an api management resource. Unfortunately, there is only documentation to import AWS resources via cli import. Is this possible with Azure too, and if so, how?

Hi everyone! I'm trying to use

pulumi import

to import some resources from Github. Do you have any solution from that error? If not, can you point me to the source code of this command so I can write a bugfix? Related to: https://github.com/pulumi/pulumi/issues/7046 https://github.com/pulumi/pulumi-aws/issues/1368

Hi Folks. I've just started using the Pulumi Go SDK. In some of the example code it uses type

pulumi.String

instead of just a regular Go

String

. What's the advantage of using the String from the

pulumi

package?

Hello people, I have a small obstacle on my project where beanstalk's ec2 instances take longer to create and I need the instances ID so I can work with their security groups and their ingress rules Is the any module native to pulumi to let it know it needs to wait (I rather no use use async if it's possible).

Hey, Question. Can we move resources between stacks using `aliases`'

stack

arg? (f.e.

aliases: [{ stack: 'old-stack' }]

) If not, what is this arg used for?

hey folks, is there any trick to run some clean up steps, like delete dynamically created files?

How does pulumi decide what version of a plugin to download?

Hi guys, could you please help me to figure out what is going on with the Custom resource? Pulumi reports that the CR is updated, but it’s not. I’ve tried to upgrade the pulumi_kubernetes to 3.12.1, but it didn’t help. Would appreciate any directions for debugging this issue.

Hi, I'm new to IaC, and I'm spinning up an EKS stack on AWS. I just want to make sure I need two separate stacks: one to manage Kubernetes and the nodes/pods and one to manage the EKS configuration on the AWS side of things. I'm also struggling to find documentation relating to deploying a service mesh using Pulumi

Is anyone else having issues running anything backed by the Pulumi backend? All my destroys/previews/ups are running forever and have to be interrupted...

Hello! I’m starting with Pulumi, and I’m on discovery mode to find out if it’s the right tool - so I’m a noobie in the subject of this solution. I’ve tried to find how to do it, but maybe I have the wrong perspective. I’m looking to codify two VPC: one in GCP and one in AWS connected with S2S VPN. I can do it individually VPCs (AWS and GCP separate Pulumi apps). What should I do if I need it in a single application? Is it the right approach, or shall I manage it separately and import it somehow later into a single app (I also can’t figure out how it would work - so not sure if it’s the right track)? Does anyone manage multi-cloud (GCP and AWS) in a single app?