pulumi-community #general

nice-cat-91582

09/26/2019, 4:44 PM

when i execute the lambda, I get this

Copy code

"errorType": "TypeError",
  "errorMessage": "Symbol.prototype [ @@toPrimitive ] requires that 'this' be a Symbol"

nice-cat-91582

09/26/2019, 4:45 PM

which is coming from some of the boilerplate code that

lambda.CallbackFunction

generates

nice-cat-91582

09/26/2019, 4:45 PM

specifically out of the last line of this:

nice-cat-91582

09/26/2019, 4:45 PM

Copy code

var __fn_prototype = {};
Object.defineProperty(__fn_prototype, "constructor", { configurable: true, writable: true, value: __f1 });
__fn.prototype = __fn_prototype;
var __fn_sym = Object.create(global.Symbol.prototype);
__fn[__fn_sym] = __fn;

nice-cat-91582

09/26/2019, 4:45 PM

anyone happen to understand what would be causing that error? I've tried it on both the Node 8 and Node 10 runtimes

early-musician-41645

09/26/2019, 5:27 PM

I'm using the

awsx

module for creating a Windows ec2 instance. Is there any way to specify that the instance should join a Microsft AD domain? The domain controller already exists as a managed Microsoft AD directory in AWS. I can get the instance to join "seamlessly" by creating it in the console: https://docs.aws.amazon.com/directoryservice/latest/admin-guide/launching_instance.html Or by logging in to the instance and configuring it: https://docs.aws.amazon.com/directoryservice/latest/admin-guide/join_windows_instance.html How can I achieve this with Pulumi?

quick-australia-29016

09/26/2019, 6:24 PM

Could you point me the way to attach pulumi to vscode debugger?

👀 1

handsome-actor-1155

09/26/2019, 6:32 PM

Quick question, can I set a pulumi config from within the program or is it CLI only?

billowy-laptop-45963

09/26/2019, 6:54 PM

what version of go are you guys using to use integration testing?

billowy-laptop-45963

09/26/2019, 7:36 PM

is there an absolute minimal example repo I could clone and

go test

to see if the tooling versions I have are ok

shy-daybreak-62210

09/26/2019, 7:55 PM

I was actually asked to implement a nightmare scenario everyone fears of IaC solutions like tf/pulumi: Destroy everything not managed. Is there a hook to destroy inverse-state?

worried-city-86458

09/26/2019, 8:42 PM

Has anybody got the prometheus-operator installing via pulumi on an eks cluster using the helm chart? I'd be interested to see your config values...

damp-book-35965

09/26/2019, 9:12 PM

Is there a way to replicate

pulumi stack ls

in typescript code currently ?

adamant-dress-73325

09/26/2019, 9:27 PM

We are seeing pulumi take 3+ minutes to read awsec2Vpc and subnets during an update? This is slowing our deployments quite a bit, any reason this might be happening?

Copy code

const vpc = awsx.ec2.Vpc.fromExistingIds("vpc", {
  privateSubnetIds: vpcPrivateSubnetIds,
  publicSubnetIds: vpcPublicSubnetIds,
  vpcId
});

brash-plastic-65797

09/26/2019, 9:52 PM

AWS suggests that you create alarms for EC2 instances which recover/reboot the instance if the status checks fail. This is easy to create through the AWS console, but having some trouble with Pulumi

worried-city-86458

09/27/2019, 12:19 AM

https://github.com/helm/charts/tree/master/stable/prometheus-operator#kubeproxy Is there an easy way to do the above with Pulumi?

worried-city-86458

09/27/2019, 12:20 AM

i.e. patch metricsBindAddress from 127.0.0.1:10249 to 0.0.0.0:10249, or more generally how to patch a k8s resource

worried-city-86458

09/27/2019, 4:57 AM

Re prometheus operator above, nearly there:

Copy code

const monitoringNamespace = new k8s.core.v1.Namespace("monitoring", { metadata: { name: "monitoring" } }, { provider : provider });

function setMonitoringNamespace(obj: any) {
    if (obj.metadata.namespace === undefined) {
        obj.metadata.namespace = "monitoring";
    }
}

const prometheusOperatorChart = new k8s.helm.v2.Chart("po", {
    repo: "stable",
    chart: "prometheus-operator",
    namespace: "monitoring",
    transformations: [ setMonitoringNamespace ],
    values: {
        kubeControllerManager: { enabled: false },
        kubeEtcd: { enabled: false },
        kubeScheduler: { enabled: false },
        kubeTargetVersionOverride: k8sVersion,
        prometheusOperator: { createCustomResource: false }
    }
}, {
    dependsOn: monitoringNamespace,
    provider: provider
});

worried-city-86458

09/27/2019, 4:57 AM

Now just seem to have a problem that the grafana-test pod fails and Pulumi waits for it until timeout

worried-city-86458

09/27/2019, 5:00 AM

pulumi-up.txt

worried-city-86458

09/27/2019, 5:01 AM

Not sure how to work around the test pod failure.

worried-city-86458

09/27/2019, 5:03 AM

Also, it would be nice to turn off warnings like:

Copy code

warning: extensions/v1beta1/PodSecurityPolicy is not supported by Kubernetes 1.16+ clusters. Use policy/v1beta1/PodSecurityPolicy instead.

limited-rainbow-51650

09/27/2019, 8:04 AM

I have a small question but more for the people who create(d) the Pulumi website. Could you tell me how, or with which tool the animated terminal images are created like in the

Create > Deploy > Manage

section of the front page?

fancy-translator-86814

09/27/2019, 9:55 AM

Hi. Today, when I run

pulumi up

, I got notification for update:

Copy code

warning: A new version of Pulumi is available. To upgrade from version '1.1.0' to '1.2.0', run
   $ brew upgrade pulumi
or visit <https://pulumi.com/docs/reference/install/> for manual instructions and release notes.

but in actual, when I run

brew upgrade pulumi

Copy code

$ brew upgrade pulumi
Error: pulumi 1.1.0 already installed

This is a bit annoying. Is this lag comes from brew? Is it possible not to show this update message?

worried-musician-21478

09/27/2019, 12:26 PM

Hello ! I'm having an issue with pulumi since today (i haven't done the 1.2.0 update). When creating an api-gateway with awsx for a new stack, it get stucks on "creating" on awsapigatewayRestApi, and in AWS a new duplicated gateway pops up every 2 minutes until I cancel the deployment... The code did not change since the last successful deployment on another stack this morning. Did you ever come across this issue ?

stale-park-1622

09/27/2019, 2:08 PM

why is the following code, creating 6 resources instead of 3? If I understood the api correctly, i can create a ECR repository and pass it into the buildAndPushImage

Copy code

export function createApiImage(name: string, port: number, contextDir: string): {repositoryUrl: pulumi.Output<string>, image: awsx.ecr.RepositoryImage} {
  
  const gitCommit = 'e0e5162';

  // set up ECR Lifecycle Policies (<https://docs.aws.amazon.com/AmazonECR/latest/userguide/LifecyclePolicies.html>) 
  // remove any images that are older than one week that are untagged
  const repository = new awsx.ecr.Repository(`pulumi-${pulumi.getStack()}-repository`, {
    lifeCyclePolicyArgs: {
      rules: [{
          selection: "untagged",
          maximumAgeLimit: 7,
      }],
    }
  });

  // And publish its URL, so we can push to it if we'd like.
  const repositoryUrl = repository.repository.repositoryUrl;

  const build: docker.DockerBuild = {
    context: contextDir, // context is a path to a directory to use for the Docker build context
    dockerfile: path.join(contextDir, './docker/api.v2.Dockerfile'), // dockerfile may be used to override the default Dockerfile name and/or location
    args: {
      EXPOSE_PORT: String(port),
    },
  };

  // Invoke 'docker' to actually build the DockerFile that is in the folder relative to
  // this program. Once built, push that image up to the ECR repo.
  //const image = repository.buildAndPushImage(build);
  const image = awsx.ecr.buildAndPushImage(
    `${name}:${gitCommit}`,
    build,
    {
      repository: repository.repository,
    }
  );

  return {repositoryUrl, image};
}

Copy code

→ pulumi up
Previewing update (dev):

     Type                           Name                          Plan
 +   pulumi:pulumi:Stack            dev                    create
 +   ├─ awsx:ecr:Repository         dev-api-image:e0e5162  create
 +   │  └─ aws:ecr:LifecyclePolicy  dev-api-image:e0e5162  create
 +   └─ awsx:ecr:Repository         pulumi-dev-repository         create
 +      ├─ aws:ecr:Repository       pulumi-dev-repository         create
 +      └─ aws:ecr:LifecyclePolicy  pulumi-dev-repository         create

Resources:
    + 6 to create

Do you want to perform this update?
  yes
> no
  details

flat-guitar-86436

09/27/2019, 5:39 PM

I’m looking to add tests around my Pulumi programs.. what’s the current status on integration testing? I found: https://github.com/pulumi/pulumi/issues/2287 https://www.pulumi.com/blog/testing-your-infrastructure-as-code-with-pulumi/#deployment-testing

stale-park-1622

09/27/2019, 5:45 PM

Is it possible to tag an image after its been build like the following suggestion from aws? https://docs.aws.amazon.com/AmazonECR/latest/userguide/docker-push-ecr-image.html The issue is, when I build the docker image using the following:

Copy code

const image = awsx.ecr.buildAndPushImage(
    `${name}:${gitCommit}`,
    build,
    {
      repository: repository.repository,
    }
  );

it always add the 'latest' tag. I want to tag it with a git commit hash. Is there a function like 'tagAndPushImage'?

damp-book-35965

09/27/2019, 5:52 PM

I'm using the awsx apigateway package to get started and it is great..But it creates a public API..is there a way to make the access limited while still using awsx. I know the pulumi-aws will support this but wanted to use awsx. Preferably unaccessible other than a certain whitelisted range of IPs rather than needing an API key to access the API ?

billowy-laptop-45963

09/27/2019, 6:30 PM

is it possible to hook into pulumi update/create/destroy events and for example make a final backup before destroy?