Is there a way to move existing infrastructure into component resources without deleting/recreating anything?

Hey all, I am having issues creating a namespace and referencing that namespace name while using the kubernetes.core.v1.Namespace (kubernetes provider). The namespace name I want is

istio-system

, but a randomly generated string gets appended to the name. Is there a way to specify a specific name for a namespace when using

new kubernetes.core.v1.Namespace

?

const istioNamespaceName = "istio-system"

const istioNamespace = new k8s.core.v1.Namespace(istioNamespaceName);

const istioVersion = "1.15.0";

const istioBase = new k8s.helm.v3.Chart("istio-base", {
    chart: "base",
    version: istioVersion,
    namespace: istioNamespaceName,
    fetchOpts: {
        repo: "<https://istio-release.storage.googleapis.com/charts>",
    },
}, {
    dependsOn: istioNamespace
});

👋 I’m trying to deploy an update to a Slack App I made. If I run

pulumi up

I get:

error modifying Lambda Function (lambda-name) Code: RequestEntityTooLargeException: status code: 413

If I run

pulumi up

again, it looks like the changes go through:

Type                    Name                                      Status      Info
     pulumi:pulumi:Stack     lambda-name                                
 ~   ├─ aws:lambda:Function  lambda-slash-commands  updated     [diff: ~code]
 ~   └─ aws:lambda:Function  lambda-cron            updated     [diff: ~code]

Resources:
    ~ 2 updated
    17 unchanged

If I then go and look for the changes in the Slack App, they’re not there. Side Note - I know I need to address the

RequestEntityTooLargeException

. I tried adding webpack to reduce the build size but that resulted in an error where Webpack doesn’t know what to do with

@pulumi/aws

and that dead-ended me there. Any thoughts on why the updates appear to successfully go through but don’t?

Hey everyone, I've been running into an issue when deploying my stacks recently. I have several projects that use Azure-Native, and recently if I run a

pulumi refresh

and then try to run

pulumi up

I have been getting this error. Running

pulumi destroy

first allows me to successfully use

pulumi up

but it is annoying to have to do this, especially with certain resources that take a long time to remove and deploy. This also doesn't seem to happen if I don't run the

refresh

before running the up command. Can anyone provide insight into what this error is and/or what might be causing it? I am using Pulumi 3.40.1. We are deploying resources to azure using the Azure-Native and pulumi standard packages. Let me know what other information I can provide in order to diagnose this issue. Thanks in advance.

Hello Pulumi masters, I am wondering if anyone of you have a recommendation on how to use a single API Gateway across multiple repos so I can keep the same base domain. I want to be able to add routes to an existing Gateway vs recreate the gateway.

Hi i ran into a problem all of a sudden and tried refreshing and and getting same result. It was working fine in previews pulumi up but all of a sudden i was getting this error when ran pulumi up or pulumi refresh or pulumi destroy. Btw i can access my cluster from same bash with kubectl get pods. i did not deleted cluster.

Hello everyone, What in your experience is the smoothest way in how we can effectively provision infrastructure across +40 accounts simultaneously. We have taken a deeper look in the documentation, but still it seems to be quite cumbersome using multiple AWS providers at the same time. It would be good if you could support/assist us in setting up this in an easy and effective manner. Thank you for your support in advance! Kind regards, Bala

Does anyone know how to enable logging in github actions?

Hello All, Azure just released Firewall Basic Sku not sure pulumi is supported this sku or not? 🙏

I’m going to try and create an open source library of dynamic providers for interacting with Bitbucket. It’ll do what the GitHub provider does but for Bitbucket since BB isn’t in the pulumi registry. I’m not finding many thorough examples of testing using python/pytest, especially nothing with dynamic resources. Do you know of an example project I could look at with tests (unit, integration, functional) for dynamic resources in python? We’ll try to make the coverage and quality good so that the community can use this library.

Hello all, after publishing NPM package of the multi-language support SDK with typescript . I am getting this error while using that npm package `error: Could not automatically download and install resource plugin 'pulumi-resource-s24'at version v0.0.0, install the plugin using

pulumi plugin install resource s24 v0.0.0

.` Underlying error: error downloading plugin s24 to file: failed to download plugin: b24-0.0.0: 403 HTTP error fetching plugin from https://get.pulumi.com/releases/plugins/pulumi-resource-s24-v0.0.0-darwin-arm64.tar.gz Thank you for your support in advance! Best regards, Bala

Hi everyone, has anyone seen this behaviour? any hints at a workaround very much appreciated. With the explicit kubernetes provider enforcement and ConfigGroup and a ComponentResource I'm getting this error:

error: TypeError: Cannot convert undefined or null to object
        at Function.keys (<anonymous>)
        at Object.transferProperties (/Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@pulumi/runtime/rpc.ts:43:28)
        at /Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@pulumi/runtime/resource.ts:504:27
        at Generator.next (<anonymous>)
        at /Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@pulumi/pulumi/runtime/resource.js:21:71
        at new Promise (<anonymous>)
        at __awaiter (/Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@pulumi/pulumi/runtime/resource.js:17:12)
        at prepareResource (/Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@pulumi/pulumi/runtime/resource.js:340:12)
        at Object.registerResource (/Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@pulumi/runtime/resource.ts:286:24)
        at new Resource (/Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@pulumi/resource.ts:401:13)
    error: Error: Invoke: Default provider for 'kubernetes' disabled. 'kubernetes:yaml:decode' must use an explicit provider.
        at Object.callback (/Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@pulumi/runtime/invoke.ts:159:33)
        at Object.onReceiveStatus (/Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@grpc/grpc-js/src/client.ts:338:26)
        at Object.onReceiveStatus (/Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@grpc/grpc-js/src/client-interceptors.ts:426:34)
        at Object.onReceiveStatus (/Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@grpc/grpc-js/src/client-interceptors.ts:389:48)
        at /Users/pprazak/repos/reflek/infrastructure-cloud/typesense/node_modules/@grpc/grpc-js/src/call-stream.ts:276:24
        at processTicksAndRejections (node:internal/process/task_queues:77:11)

This is the relevant fragment of the component:

const glbConfig = pulumi.all([args.namespace, this.internalHostname]).apply(
            ([ns, hs]) => new kubernetes.yaml.ConfigGroup(
                `${publicServiceName}-glb-config`,
                {
                    transformations: [args => {
                        return {
                            // failed attempt #1 at a workaround
                            opts: pulumi.mergeOptions(args.opts, {parent: this})
                        }
                    }],
                    yaml: [
                        `apiVersion: <http://cloud.google.com/v1|cloud.google.com/v1>
kind: BackendConfig
metadata:
  name: ${publicServiceName}
  namespace: ${ns}
spec:
  healthCheck:
    checkIntervalSec: 15
    timeoutSec: 15
    healthyThreshold: 1
    unhealthyThreshold: 2
    type: HTTP
    requestPath: /health
    port: ${args.apiPort || defaultApiPort}
`, `
apiVersion: <http://networking.gke.io/v1beta1|networking.gke.io/v1beta1>
kind: FrontendConfig
metadata:
  name: ${publicServiceName}
  namespace: ${ns}
spec:
  redirectToHttps:
    enabled: true
    responseCodeName: MOVED_PERMANENTLY_DEFAULT
`]
                }, {
                    parent: this,
                    transformations: [
                        (args: ResourceTransformationArgs) => {
                            return {
                                // failed attempt #2 at a workaround
                                opts: pulumi.mergeOptions(args.opts, {parent: this})
                            } as ResourceTransformationResult
                        }
                    ],
                }
            ));

I'm setting the provider explicitly on the component:

providers: {
        kubernetes: k8sProvider, // make sure we use explicit Kubernetes provider, not the default
    },

Looks like the provider set on the component is not propagating to the Yaml.

Hi. trying to deploy the sample for Cloudfront distrib with Cloudfront origin control with latest pulumi.

example = aws.cloudfront.OriginAccessControl("example",
    description="Example Policy",
    origin_access_control_origin_type="s3",
    signing_behavior="always",
    signing_protocol="sigv4")

Getting this error message, any fix available?

AttributeError: module 'pulumi_aws.cloudfront' has no attribute 'OriginAccessControl'

Hi Team, Need some suggestions. We deploy infrastructure for users of our azure data platform. As part of infrastructure, we also create azure data factories. The users are responsible for creating their own adf data pipelines. The issue I am facing is that if we have to do redeployment of the adf then the global parameters which the users have deployed are removed. How can I fetch the latest global parameters if any, in the pulumi code.

Hi, for folks that have used the micro-stack architecture and outputs/stack references to communicate dependent data between stacks, how have you found the coupling between stacks to be using those stack references? How tightly coupled do those stacks become? If you make changes in an upstream stack (say the VPC stack), when do those changes impact the downstream stacks (say the DB stack)?

Hey all, I am currently learning Pulumi and have created my first stack. Is there documentation somewhere for making this stack reusable? Example: I want to define my infrastructure and then reuse this infrastructure to spin up different environments (qa, staging, prod, etc). What is the best approach for this scenario? Thanks!

any idea why this is happening?

I thought pulumi/actions@v3 includes the pulumi cli

Hi, I'm a total beginner. Pulumi sounds awesome! I love that I can use my favorite language (Go). I looked through https://www.pulumi.com/registry/ and it seems like something is missing. I want a package that abstracts the provider. So the Pulumi developer doesn't need to know anything about the concrete provider. AWS, Azure, and GCP (etc.) are all different, but there are many cloud resources from cloud each provider that are similar enough to be interchangeable. For example if I want to provision a new object store (e.g. AWS Simple Storage Service, Azure Blob Storage, GCP Cloud Storage) First, my code looks like:

pulumi.Run(func(ctx *pulumi.Context) error {
        // Create an object store resource
        bucket, err := abstractProvider.NewObjectStore(ctx, "my-object-store", nil)
        if err != nil {
            return err
        }
})

Then, there is a separate layer which determines which provider to use or to use multiple providers for multi-cloud redundancy. (Example selection criteria: price, data center geography, real time service outages) Does such a thing exist? Essentially, why should a Pulumi developer need to hard code the provider into the IaC? Also, looking at https://www.pulumi.com/blog/tag/native-providers/ it seems there has been no blog posts about native providers since the original blog post on Apr 20, 2021. Is this a sign that the native providers are not being prioritized?

This is a duplicate thinking taco https://pulumi-community.slack.com/archives/C01PF3E1B8V/p1665246121533419

Hi All, has anyone been able to get cert-manager installed in their k8s cluster via pulumi (https://www.pulumi.com/registry/packages/kubernetes-cert-manager/). I know just installing the helm chart doesnt work so i tried to use the pulumi resource, but on my mac it complains that it needs v0.0.3 and is on v0.0.1. My tsConfig is showing the v0.0.3 dependency and i just tried to manually update all dependencies as well but it is still failing. Is there a away around this that doesnt require manual action so my deployments can continue to be automated?

We're having an issue using AWSX where we're trying to manage a fargate service, and on every

pulumi up

it thinks we need to delete/re-create the container defn. Here is one of the diff outputs (edited). We compared the diffs across different successful runs of

pulumi up

and they are always the same.

[id=pganalyze-fargate]
    [urn=urn:pulumi:production::rainbow-infra::awsx:x:ecs:FargateTaskDefinition$aws:ecs/taskDefinition:TaskDefinition::production-pganalyze-service]
    __defaults             : []
    containerDefinitions   : "[{"image":"<http://quay.io/pganalyze/collector:stable|quay.io/pganalyze/collector:stable>","cpu":256,"memory":512,"essential":true,"ulimits":[{"name":"nofile","softLimit":32000,"hardLimit":65536}],"environment":[{"name":"DB_HOST","value":"snip},{"name":"DB_USERNAME","value":"pganalyze"},{"name":"DB_NAME","value":"rainbow"}],"secrets":[{"name":"PGA_API_KEY","valueFrom":"snip"},{"name":"DB_PASSWORD","valueFrom":"snip"}],"logConfiguration":{"logDriver":"awsfirelens"},"readonlyRootFilesystem":false,"mountPoints":[],"name":"pganalyze"},{"cpu":128,"memory":512,"image":"<http://XXX.dkr.ecr.us-west-2.amazonaws.com/amazon/aws-for-fluent-bit:2.26.0|XXX.dkr.ecr.us-west-2.amazonaws.com/amazon/aws-for-fluent-bit:2.26.0>","firelensConfiguration":{"type":"fluentbit","options":{"config-file-type":"file","config-file-value":"/log_destinations.conf"}},"environment":[{"name":"SERVICE_NAME","value":"pganalyze"},{"name":"SERVICE_ENV","value":"production"},{"name":"GRAFANA_USER","value":"XXX"},{"name":"GRAFANA_API_KEY","value":"[secret]"}],"essential":true,"logConfiguration":{"logDriver":"awslogs","options":{"awslogs-group":"/ecs/ecs-aws-fluentbit-sidecar-logger","awslogs-region":"us-west-2","awslogs-stream-prefix":"pganalyze","awslogs-create-group":"True"}},"name":"logs"}]"
    cpu                    : "512"
    executionRoleArn       : "arn:aws:iam::XXX:role/rainbow-pganalyze-role-2961ff0"
    family                 : "pganalyze-fargate"
    memory                 : "1024"
    networkMode            : "awsvpc"
    requiresCompatibilities: [
        [0]: "FARGATE"
    ]
    taskRoleArn            : "arn:aws:iam::XXX:role/rainbow-pganalyze-task-role-919c4b6"%

Any ideas? What are we doing wrong?

Does anyone have any good examples of using Pulumi for a serverless project? I know there are a few getting started examples, but I'm looking for more info about how to structure projects, connect functions and what not. Something more "real world".

We've just received a notification from GitHub that the Pulumi GitHub App now requires additional permissions to environments and deployments. However, we couldn't find any documentation on new features that would warrant these additional permissions. Could someone from the @echoing-match-29901 elaborate on this, please? Thanks! 🙏

Hey Team 👋 I'm new to the slack workspace and have a technical question I'd like to ask. What is the best channel for this purpose?

How can i speed up docker build on CI/CD? I am using docker.RemoteImage

Hi all, I have a structure like :

└── common
    ├── Pulumi.cmn.yaml
    ├── Pulumi.yaml
    ├── __main__.py
    ├── __pycache__
    ├── kubernetes
    │   ├── Pulumi.kubernetes.yaml
    │   ├── Pulumi.yaml
    │   ├── __main__.py
    │   ├── __pycache__
    │   └── requirements.txt
    └── requirements.txt

I would like to have micro stacks and use the GCP bucket as the backend. I’m pointing to the same GCP bucket in both Pulumi.yaml files. Here is stacks list:

pulumi stack ls
NAME        LAST UPDATE     RESOURCE COUNT
cmn         12 minutes ago  3
kubernetes  22 hours ago    0

in kubernetes

__main__.py

file I have this part of code

stack = pulumi.get_stack()
print(stack)
pulumi.stack_reference(f"cmn")

and I’m getting the error:

pulumi:pulumi:Stack (kubernetes-kubernetes):
    error: Program failed with an unhandled exception:
    Traceback (most recent call last):
      File "/Users/irmas/projects/devops-new/infra/common/kubernetes/./__main__.py", line 16, in <module>
        pulumi.stack_reference("cmn")
    TypeError: 'module' object is not callable
 
    kubernetes

@shy-arm-32391 are you taking over the work from Paul Stack? https://github.com/pulumi/pulumi-gitlab/issues/183 this contains an update that is critical for my work, would it be possible to merge it?

Hello , everyone ， I would like to know how to use

OutPut.unsecret

, I found some test case in

sdk/python/lib/test/test_next_serialize.py

.

class OutputSecretTests(unittest.TestCase):
    @pulumi_test
    async def test_secret(self):
        x = Output.secret("foo")
        is_secret = await x.is_secret()
        self.assertTrue(is_secret)

    @pulumi_test
    async def test_unsecret(self):
        x = Output.secret("foo")
        x_is_secret = await x.is_secret()
        self.assertTrue(x_is_secret)

        y = Output.unsecret(x)
        y_val = await y.future()
        y_is_secret = await y.is_secret()
        self.assertEqual(y_val, "foo")
        self.assertFalse(y_is_secret)

but there are not async function in my pulumi script , I’m not using asynchronous, I don’t know how to use this.