pulumi-community #general

Docs

Join the conversation Join Slack

general

proud-spoon-58287

08/03/2020, 4:39 PM

got a the clusterprovider

proud-spoon-58287

08/03/2020, 4:39 PM

do not get how to create a link between the two things

average-eye-3295

08/04/2020, 12:18 AM

Hi everyone!.Guys. I have this issue. I am triying to deploy from Jenkins in DEV environment but I have the follow error 409. Can you help me?

docker:image:Image dev-data-gathering-srv-docker-img  The push refers to repository [<http://gcr.io/pivotal-spark-262418/dev-data-gathering-srv|gcr.io/pivotal-spark-262418/dev-data-gathering-srv>]
    docker:image:Image dev-data-gathering-srv-docker-img  Successfully pushed to docker.
 ~  gcp:cloudrun:Service dev-data-gathering-srv updating [diff: ~template]
 ~  gcp:cloudrun:Service dev-data-gathering-srv updating [diff: ~template]; error: 1 error occurred:
 ~  gcp:cloudrun:Service dev-data-gathering-srv **updating failed** [diff: ~template]; error: 1 error occurred:
    pulumi:pulumi:Stack cloud-run-deployer-dev running error: update failed
    pulumi:pulumi:Stack cloud-run-deployer-dev **failed** 1 error
 
Diagnostics:
  pulumi:pulumi:Stack (cloud-run-deployer-dev):
    error: update failed
 
  gcp:cloudrun:Service (dev-data-gathering-srv):
    error: 1 error occurred:
    	* updating urn:pulumi:dev::cloud-run-deployer::gcp:cloudrun/service:Service::dev-data-gathering-srv: Error updating Service "locations/us-central1/namespaces/pivotal-spark-262418/services/dev-data-gathering-srv-2a3d6f8": googleapi: Error 409: Conflict for resource 'dev-data-gathering-srv-2a3d6f8': version '1594441718486000' was specified but current version is '1594441734108000'.
 
Outputs:
  - Domain               : "<http://dev.data-gathering.srv.yofio.app|dev.data-gathering.srv.yofio.app>"
  - dockerImg            : "<http://gcr.io/pivotal-spark-262418/dev-data-gathering-srv:latest-XXSXSXS12313131|gcr.io/pivotal-spark-262418/dev-data-gathering-srv:latest-XXSXSXS12313131>"
  - service account Email: "dev-data-gathering-svr-cr@pivotal-spark-262418.iam.gserv

average-receptionist-65239

08/04/2020, 6:21 AM

What do folks call their Pulumi TypeScript files? Pulumi? Pulumi config/configuration? Infrastructure? IaC? Infrastructure definition? i.e. "I'm gonna add

kube-state-metrics

to the Pulumi configs"

proud-spoon-58287

08/04/2020, 1:02 PM

Hi all, I got the current situation: a kubernetes yaml containing two secrets that obviously cannot be checked in Git. I have then created two keys in gcp secret service. Now my question is: how do I use Pulumi to fetch those values and replace them inside the yaml file before being deployed into my kubernetes cluster?

gifted-student-18589

08/04/2020, 1:59 PM

quick question - if my projects needs staging/prod/xyz, should these be separate Pulumi stacks? What’s a good practice? A link would be appreciated as well. Thank you in advance! 🍻

delightful-xylophone-3967

08/04/2020, 3:22 PM

Dear Pulumi Team, is anybody working on issue #4224 (https://github.com/pulumi/pulumi/issues/4224)? We would love to deploy the certmanager CRDs over pulumi, but it seems that it is currently impossible with the c# yaml parser? We are facing the same problem, as described in the issue.

boundless-musician-43495

08/04/2020, 5:55 PM

Hi all - I'm not sure if this is possible, or if i'm just going about it the wrong way I'm using a dotnet project for deployment. Locally, letting it run the build and find the binary is fine, but on the build server I want to be able to specify a path to a binary I can't find a way to override the runtime properties using a stack config file. Am I going about this completely wrong?

gray-city-50684

08/04/2020, 8:06 PM

I ran into a situation, where when I create a Kubernetes cluster (using Pulumi) the returned cluster still deploys some components (Linkerd2 in this case) when my Pulumi code continues with deploying my resources (ingress controller, echo server - this is for a demo)

astonishing-quill-88807

08/04/2020, 8:18 PM

For configuring providers using the stack settings, do all of the keys need to be of the form

vault:address

vault:auth_logins

, or can you do something like:

vault:
    address: foo
    auth_logins:
        - bar

average-receptionist-65239

08/05/2020, 5:33 AM

I use

<http://pulumi.log.info|pulumi.log.info>()

and

pulumi.log.debug()

in my TS infrastructure code. How can I see the debug logging during a preview? I've tried

pulumi preview --verbose 3

(and 4) but that's not it.

worried-sandwich-8337

08/05/2020, 5:41 AM

I need to pass a secret from AWS Secret Manager to a container using an array of

aws.ecs.Secret

. The problem is

valueFrom

is a

String

but I only have a

Promise<string>

which can be converted to

Output<string>

via

pulumi.output

. What's the recommended solution? Thanks! https://github.com/pulumi/pulumi-aws/blob/master/sdk/nodejs/ecs/container.ts#L262

bumpy-river-47200

08/05/2020, 6:19 AM

Hi, using gcp dataflow python in pulumi. I am using standard package as mentioned below but somehow it always going into error. My Job is template based job, Which refers to a metadata file which has run time parameters (and it seems pulumi unable to parse those runtime parameters while executing).

*Code*:

dfjob = gcp.dataflow.Job(
  "mydfjob",
    max_workers=4, 
    name="mydfjob", 
    parameters={ 
      "subscriber": "projects/demo/subscriptions/demo_subs",
      "CustomSize": "2",
      "callingAgent": "py_data_hourly"
    },
    project="demo", 
    region="us-west1", 
    temp_gcs_location="<gs://demotemplates/tempfiles>",
    template_gcs_path="<gs://demotemplates/templates/mydfjob.json>",
    zone="us-west1b", 
    )

Error:

Diagnostics:
  gcp:dataflow:Job (mydfjob):
    error: googleapi: Error 400: (5e3de3139df6c08e69): The workflow could not be created. Causes: (5ee31t539df6c08f66): Found unexpected parameters: ['callingAgent' (perhaps you meant 'runner'), 'CustomSize' (perhaps you meant 'runner'), 'subscriber' (perhaps you meant 'pipelineUrl')], badRequest

Any suggestion to fix this?

👀 1

kind-mechanic-53546

08/05/2020, 9:03 AM

Warning for anyone upgrading to 2.8.0 Broken upgrade to 2.8.0 for passphrase secrets provider #5104 Only tested with passphrase secrets provider, possible other providers are unaffected

mammoth-motorcycle-8004

08/05/2020, 9:24 AM

Hi, got a problem creating an EKS stack with Pulumi. The Nodegroup members are not joining the cluster, so "Kubentl get nodes" shows nothing. It seems to be because the latest AMIs for Kubernetes (https://docs.aws.amazon.com/eks/latest/userguide/eks-optimized-ami.html) are not compatible with the Node Template being used in Cloudformation. I say this because if I pass in an old AMI ID the cluster is created OK and the Node Group members join the cluster. In the link it says the CloudFormation template needs to be correct or this will happen... Fairly new to Pulumi, I'm using the latest version of pulumi/eks - how do I get this to work?

nutritious-shampoo-16116

08/05/2020, 10:34 AM

hello everybody, is there anyone who's using pulumi+ci to deploy docker containers using ECS? If so how you manage versions to be deployed?

proud-spoon-58287

08/05/2020, 11:00 AM

Hi all, I have a configMap and I wonder how I can pass that map in the pod env.

nutritious-battery-42762

08/05/2020, 12:57 PM

i want to auto update godaddy records but the only solution i found is https://github.com/baronfel/pulumi-godaddy but it's not working ,any suggestion?

astonishing-quill-88807

08/05/2020, 4:31 PM

So, I've got a repository with multiple different modules that each have a

__main__.py

and I'm creating separate stacks for each of them using an S3 backend. Currently I've been creating the stack definitions inside the modules, but I accidentally ended up with one at the root of the project and it automatically added a line to the config with a path to the module. I just wanted to double check whether my initial approach is the "correct" way, or if I should be creating the stack definition yaml in the root of the repository? Repository: https://github.com/mitodl/ol-infrastructure/ Root level stack definition: https://github.com/mitodl/ol-infrastructure/blob/main/Pulumi.infrastructure.aws.dns.yaml Module level stack definition: https://github.com/mitodl/ol-infrastructure/blob/main/src/ol_infrastructure/infrastructure/aws/dns/Pulumi.infrastructure.aws.dns.yaml

astonishing-quill-88807

08/05/2020, 8:34 PM

I have a stack that's using an S3 backend and KMS encryption which I recently renamed, and now Pulumi is trying to get me to enter my passphrase when I attempt to run an

up

command. Is there something I need to do to get it to recognize the fact that I've already got the KMS secret management set up?

many-egg-6856

08/05/2020, 8:46 PM

Does setting a backend URL in

pulumi.yaml

override the backend set by

pulumi login

proud-spoon-58287

08/05/2020, 9:48 PM

Hi all, I wonder why the following

pulumi.interpolate`"org.apache.kafka.common.security.plain.PlainLoginModule required username=\\"${config.requireSecret('kafkaApiKey')}\\" password=\\"${config.requireSecret('kafkaApiSecret')}\\";"`

thankful-holiday-28161

08/05/2020, 10:53 PM

Is there a way in order to drop into a debugger statement within the Pulumi config? Either

import ipdb; ipdb.set_trace()

for Python or

debugger;

for TypeScript?

thankful-holiday-28161

08/05/2020, 10:56 PM

Also I think the Markdown for the Pulumi documentation isn't parsed correctly for at least this page: https://www.pulumi.com/docs/reference/pkg/nodejs/pulumi/gcp/compute/#BackendBucketSignedUrlKey; this is what I see

thankful-holiday-28161

08/05/2020, 10:56 PM

A key for signing Cloud CDN signed URLs for BackendBuckets. To get more information about BackendBucketSignedUrlKey, see: * [API documentation](<https://cloud.google.com/compute/docs/reference/rest/v1/backendBuckets>) * How-to Guides * [Using Signed URLs](<https://cloud.google.com/cdn/docs/using-signed-urls/>) > **Warning:** All arguments including `keyValue` will be stored in the raw state as plain-text. #### Example Usage - Backend Bucket Signed Url Key ```typescript import * as pulumi from "@pulumi/pulumi"; import * as gcp from "@pulumi/gcp"; const bucket = new gcp.storage.Bucket("bucket", {location: "EU"}); const testBackend = new gcp.compute.BackendBucket("testBackend", { description: "Contains beautiful images", bucketName: bucket.name, enableCdn: true, }); const backendKey = new gcp.compute.BackendBucketSignedUrlKey("backendKey", { keyValue: "pPsVemX8GM46QVeezid6Rw==", backendBucket: testBackend.name, }); ```

damp-elephant-82829

08/06/2020, 7:51 AM

How does one wait an awaitable resource in python?!

astonishing-quill-88807

08/06/2020, 1:27 PM

I'm trying to debug why a set of resources are set to be deleted and recreated despite there not being any apparent changes. Any pointers?

damp-elephant-82829

08/06/2020, 2:07 PM

@astonishing-quill-88807 I would look what happens if you do pulumi refresh. If that doesn’t help, I would enable terraform tracing

damp-elephant-82829

08/06/2020, 2:07 PM

export TF_TRACE=TRUE

astonishing-quill-88807

08/06/2020, 2:07 PM

Great, thanks