pulumi-community #google-cloud

Docs

Join the conversation Join Slack

google-cloud

boundless-intern-43214

04/29/2021, 10:01 AM

But I receive the following error:

boundless-intern-43214

04/29/2021, 10:01 AM

error: gcp:container/cluster:Cluster resource 'cluster' has a problem: parsing time "07:00" as "2006-01-02T15:04:05Z07:00": cannot parse "0" as "2006". Examine values at 'Cluster.MaintenancePolicy.RecurringWindow[0].StartTime'.
    error: gcp:container/cluster:Cluster resource 'cluster' has a problem: parsing time "15:00" as "2006-01-02T15:04:05Z07:00": cannot parse "0" as "2006". Examine values at 'Cluster.MaintenancePolicy.RecurringWindow[0].EndTime'.

boundless-intern-43214

04/29/2021, 10:01 AM

What is the correct format to use in the

startTime

and

endTime

wide-glass-98978

04/29/2021, 7:42 PM

so it appears that when I

destroy

projects.NewIAMBinding

that there is collateral damage and all other bindings with the same role are getting nuked from the target projects

wide-glass-98978

04/29/2021, 7:44 PM

for example, I have a project with a GCR registry, and I'm adding IAM bindings for other projects who need to read from that registry (

storage.objectViewer

), but when I delete one of the reader stacks, and clobbers all of the bindings in the project that grant

storage.objectViewer

wide-glass-98978

04/30/2021, 2:16 PM

Figured it out, for anyone who sees this, I was using

IAMBinding

and I should have been using

IAMMember

IAMBinding

sets the entire project membership for the named role, which is very surprising as someone familiar with k8s `rolebinding`s

red-area-47037

05/01/2021, 2:24 PM

Hi, quick question, I wanted to test Googles Dataplane V2 and adjust my Pulumi script to create a cluster with Dataplane v2 enabled.. I checked the documentation (I checked the "classic" and the new "native" provider), but I couldn't find the needed config option 😞

modern-animal-54928

05/02/2021, 5:40 PM

Hello everyone

modern-animal-54928

05/02/2021, 5:40 PM

I am facing some weird issue

modern-animal-54928

05/02/2021, 5:41 PM

in a single program I'm trying to create some projects then to add iammember and some API

modern-animal-54928

05/02/2021, 5:42 PM

but it is failing because the project seems not to be created while the iammember and the services are added

modern-animal-54928

05/02/2021, 5:42 PM

How do you manage that ?

modern-animal-54928

05/02/2021, 9:37 PM

Ok I found it. I have used the depends_on

modern-animal-54928

05/02/2021, 9:49 PM

Meanwhile I am facing a strange behaviour : I am creating a project, and I am trying to launch an vm inside. But everytime I get an error about a missing API

Error creating Address: googleapi: Error 403: Compute Engine API has not been used in project XXX before or it is disabled. Enable it by visiting <https://console.developers.google.com/apis/api/compute.googleapis.com/overview?project=XXX> then retry.

Can you confirm me that adding some services to that project is enough ? I am adding the following ones:

services=["<http://cloudapis.googleapis.com|cloudapis.googleapis.com>", "<http://compute.googleapis.com|compute.googleapis.com>",  "<http://logging.googleapis.com|logging.googleapis.com>", "monitori    <http://ng.googleapis.com|ng.googleapis.com>","<http://cloudbilling.googleapis.com|cloudbilling.googleapis.com>"]

Thanks !

most-magician-78411

05/04/2021, 12:17 AM

Hello everyone! can someone confirm if there's is or isn't a way to manage Cloud SQL postgres replicas from Pulumi? (I looked in the docs and tried searching but there doesn't seem to be a way). Thanks!

most-magician-78411

05/04/2021, 3:51 AM

is there a way to work around import errors like when trying to import a postgress replica? any help would be much appreciated. thanks!

proud-pizza-80589

05/12/2021, 9:06 PM

Moving from the old gcp to the new native one, can you just adjust everything or will it go and recreate everything?

cuddly-father-4905

05/12/2021, 11:42 PM

I have multiple GCP projects, one for each environment (dev, QA, staging, production, etc) Is there a best practice for how those projects should be created in the first place? Should I manually create a separate project through the GCP console with a service account that has permission to create projects and then use that? Should that same "master" service account also be used for all Pulumi operations (i.e. for creating all resources in all projects), or should I create a service account in each separate project and use those for Pulumi?

quiet-truck-58459

05/17/2021, 9:41 PM

Hey. Has anyone managed to run a cloudbuild.Build (not a Trigger) ?

quiet-truck-58459

05/19/2021, 11:49 AM

Is anyone using

pulumi_google_native

with Python?

adorable-continent-4073

05/31/2021, 10:23 PM

hello! can anyone point me to a working example of accessing a GCP secret? When I look at the docs it's asking for

google-beta

provider, but erroring out on that (Typescript). I'm working from the example here. the error i'm getting is this:

TSError: ⨯ Unable to compile TypeScript:
    index.ts(84,19): error TS1136: Property assignment expected.
    index.ts(65,13): error TS2304: Cannot find name 'google_beta'.

quick-painter-61380

06/02/2021, 2:27 PM

what's the difference between how the old gcp provider and the new one work? I assume from the name that the new one is making direct gRPC calls or something but not sure how the old one works? I only ask because it seems the native one is still considered beta.

adorable-continent-4073

06/02/2021, 10:22 PM

anybody have any suggestions on how to debug

resource does not exist

from

pulumi import

? I can see the resource (a cloudrun service) in console, and I'm providing the name to

pulumi import

like so:

pulumi import gcp:cloudrun/service:Service ruby-service uswest-1/pulumi-training-xxx/ruby-xyzw

where

pulumi-training-xxx

is my GCP project ID, and

ruby-xyzw

is my cloud run service ID (which i created with pulumi, actually). I've tried all the format variations for

pulumi import

and nothing is ever found by pulumi.

clever-byte-21551

06/03/2021, 5:32 AM

Did anybody try to destroy a stack containing a CryptoKey? I’m unable to do that AT ALL, the object is in “destroyed” state in GCP but it’s never deleted (from my understanding it can’t be deleted) https://github.com/pulumi/pulumi-gcp/issues/587

gifted-planet-66368

06/09/2021, 11:06 AM

I just setup my first test-project and am facing this error message on MacOS:

-> % pulumi up      
Enter your passphrase to unlock config/secrets
    (set PULUMI_CONFIG_PASSPHRASE or PULUMI_CONFIG_PASSPHRASE_FILE to remember): 
Previewing update (dev):
     Type                 Name        Plan     Info
     pulumi:pulumi:Stack  pubsub-dev           1 error
 
Diagnostics:
  pulumi:pulumi:Stack (pubsub-dev):
    error: Running program '/Users/svs/IdeaProjects/sandbox/pulumi' failed with an unhandled exception:
    Error: 13 INTERNAL: Received RST_STREAM with code 2 triggered by internal client error: Protocol error
        at Object.callErrorFromStatus (/Users/svs/IdeaProjects/sandbox/pulumi/node_modules/@grpc/grpc-js/src/call.ts:81:24)
        at Object.onReceiveStatus (/Users/svs/IdeaProjects/sandbox/pulumi/node_modules/@grpc/grpc-js/src/client.ts:338:36)
        at Object.onReceiveStatus (/Users/svs/IdeaProjects/sandbox/pulumi/node_modules/@grpc/grpc-js/src/client-interceptors.ts:426:34)
        at Object.onReceiveStatus (/Users/svs/IdeaProjects/sandbox/pulumi/node_modules/@grpc/grpc-js/src/client-interceptors.ts:389:48)
        at /Users/svs/IdeaProjects/sandbox/pulumi/node_modules/@grpc/grpc-js/src/call-stream.ts:276:24
        at processTicksAndRejections (node:internal/process/task_queues:78:11)

I intalled Pulumi via homebrew (3.4.0). grpc-js is at the latest version (1.3.2), node version is 16.3.0

alert-helicopter-26131

06/10/2021, 7:17 AM

Tried changing the set of domains on a

gcp.compute.ManagedSslCertificate

. Got some kind of dependency error on `pulumi up`:

error: deleting urn:pulumi:staging::REDACTED::gcp:compute/managedSslCertificate:ManagedSslCertificate::app: 1 error occurred:
	* Error when reading or editing ManagedSslCertificate: googleapi: Error 400: The ssl_certificate resource 'projects/REDACTED/global/sslCertificates/app-22f6675' is already being used by 'projects/REDACTED/global/targetHttpsProxies/app-b28855a', resourceInUseByAnotherResource

I'm stuck and not sure how to get unstuck. (I tried doing

pulumi up -r

, same result.)

limited-rainbow-51650

06/21/2021, 10:56 AM

Hello, on GCP, we have GKE clusters with and without (legacy) Workload Identity enabled. We want to parameterize our code for this, but what is the most suitable way to find out if Workload Identity is enabled or not? Is

getCluster

my best option? https://www.pulumi.com/docs/reference/pkg/gcp/container/getcluster/

adorable-action-51248

06/21/2021, 6:11 PM

Hello, I switched from gcp to cloud-native for deploying cloudrun services. Somehow, when inspecting the service properties of a few services, it turns out that some were missing the status property. it seems this has something to do with the transition from gcp to cloud-native. the first deployment of cloud-native removes the old gcp service and recreates it. however, it seems that if the changes between the gcp and cloud-native versions are too small the status property is not returned anymore. when i added a new environment variable to the container definition the status property appeared. after one deployment i was able to remove the variable again and the status property remained.

eager-thailand-33661

06/24/2021, 10:10 AM

Hi, anyone tried gcp-native with Python? I get errors telling me that the semver has illegal characters when trying to install either as dependency in requirements.txt or by simply running the pulumi plugin install command.

rhythmic-actor-14991

06/25/2021, 2:18 AM