Why does this command fail when the urn exists in my stack's state file?

pulumi state delete urn:pulumi:dev::vm-scaler::kubernetes:<http://helm.sh/v2:Chart$kubernetes:rbac.authorization.k8s.io/v1:ClusterRoleBinding::dd-agent-datadog|helm.sh/v2:Chart$kubernetes:rbac.authorization.k8s.io/v1:ClusterRoleBinding::dd-agent-datadog>
 warning: This command will edit your stack's state directly. Confirm? Yes
error: No such resource "urn:pulumi:dev::vm-scaler::kubernetes:<http://helm.sh/v2:Chart:rbac.authorization.k8s.io/v1:ClusterRoleBinding::dd-agent-datadog|helm.sh/v2:Chart:rbac.authorization.k8s.io/v1:ClusterRoleBinding::dd-agent-datadog>" exists in the current state

If I search for

urn:pulumi:dev::vm-scaler::kubernetes:<http://helm.sh/v2:chart:rbac.authorization.k8s.io/v1:ClusterRoleBinding::dd-agent-datadog|helm.sh/v2:chart:rbac.authorization.k8s.io/v1:ClusterRoleBinding::dd-agent-datadog>

in my stack's json file, I will find it.

Is there a way to just delete resources from the pulumi state? Going through a deleting every dependent URN a resource may create is incredibly tedious.

Hey! Could you take a look at https://github.com/avolkov-dev/pulumi-mount-issue ? i'm getting weird error (

VolumeInUse

) when reaattaching existed volume to new instance

I am planning to deploy dev and prod stacks to separate aws accounts. The environments will also have separate stacks. What would be the best way to enforce that dev stacks (there could be many of these) go to the dev account and that the prod stack (of which there should be at most one) goes to the prod aws account? The only thing I can think of is using policy packs. In terraform you would achieve this by having separate provider files for example. Is anybody else doing this in a better way? Or am I wrong in the entire approach?

Hi! It was great seeing the team at Kubecon. Speaking of, I’m curious what thoughts anyone has around where pulumi fits in with kubernetes operators. Will pulumi just be a tool for applying the initial resources of the operator itself, and the corresponding CRDs, or does pulumi potentially have a place in building operators themselves?

hey everyone, I've got a question about turning outputs into concrete values (I think)

I'm trying to get secrets into a fargate container definition

the pulumi.aws.Secret type has name and valueFrom (ARN), both strings

I've got outputs from the parameter store secrets I've created that I would like to pass in

but since valueFrom is a string and doesn't accept an output, I'm not sure how to construct the config so it works

abbreviated config:

const parameter = new aws.ssm.Parameter("SECRET", {
    type: "SecureString",
    value: config.requireSecret("myPulumiSecret"),
});

const service = new awsx.ecs.FargateService("service", {
    cluster,
    desiredCount: 2,
    taskDefinitionArgs: {
        container: {
            secrets: [{
                name: parameter.name,
                valueFrom: parameter.arn, // FAILS, Type 'Output<string>' is not assignable to type 'string'.
            }],
        },
    }
});

I also tried the pattern shown here (https://stackoverflow.com/questions/58341593/pulumi-how-to-serialize-outputstring-to-json/58347396#58347396) with no luck

We use cloud backed storage (s3) instead of pulumi.com, can we still use

pulumi.StackReference

?

Hi all 👋 when you replace a DynamoDB table (for example when the hashkey attribute changes), will the data be deleted?

Hi all, I moved around the local directory in which our pulumi stack was configured, and this has somehow messed up the state management, despite us using a remote stack backend in a Google Cloud Storage bucket. Running any pulumi command now gives the error

Could not get signed url for stack location

. Pulumi version installed on my mac is v1.6.1 Has anybody encountered such an error before or could give pointers towards recovering from this error? I'm willing to download the state file from the storage bucket and restore the state manually, but can't find any clear directions on how to do so, thanks.

pulumi login --cloud-url <gs://xxx-yyy-zzz>
Logged into Agrim-Macbook-Pro.local as agrim (<gs://xxx-yyy-zzz>)

pulumi refresh
Enter your passphrase to unlock config/secrets
    (set PULUMI_CONFIG_PASSPHRASE to remember):
Previewing refresh (experiments):


Resources:

Do you want to perform this refresh?
No resources will be modified as part of this refresh; just your stack's state will be. yes
Refreshing (experiments):

error: Could not get signed url for stack location: blob (key ".pulumi/stacks/experiments.json") (code=Unknown): to use SignedURL, you must call OpenBucket with a valid Options.GoogleAccessID and exactly one of Options.PrivateKey or Options.SignBytes

I also tried to download a checkpoint file from the storage bucket manually, and then tried to import the stack from this file, but this command also fails:

pulumi stack import --file ~/Downloads/pulumi_history_experiments_experiments-1574820036002431000.checkpoint.json

error: could not deserialize deployment: unexpected end of JSON input

Is there a clean way to ask pulumi cli to give up? Ctrl-C causes pulumi to leave things in pending. Example: typo in k8s namespace. Pulumi waits a while and tries several times, but it is clear on the first failure that it should be stopped.

Hello Peers, I've got a beginner question: How can I create

Deplyments

,

Services

etc. in a specific

namespace

? Hope this isn't to basic of a question

Hi Guys,

We faced an issue regarding to creation of Azure App Service Environment using Pulumi. As Pulumi doesn't have direct support for ASE we aimed to create that using TemplateDeployment.

The deployment is time consuming but after couple of try, we've got bearer token error which we think it's due to using Azure CLI login and not service principal. Here is the link to the issue https://github.com/pulumi/pulumi/issues/2644

I'm wondering if anyone had the same issue before which can help

Hey friends,

import * as k8s from "@pulumi/kubernetes";
import * as pulumi from "@pulumi/pulumi";
const env = pulumi.getStack();
const infra_stack = new pulumi.StackReference(`acmecorp/infra/${env}`);
const provider = new k8s.Provider("k8s", { kubeconfig: infra.getOutput("kubeConfig") });
const service = new k8s.core.v1.Service(..., { provider: provider });

If infra_stack is deployed in aws account A and current stack is deployed in aws account B, does the code snippet still work?🙂

Is it possible to enforce pulumi to stop ec2 instance before detaching the volume on

userdata

change(and instance recreation)? https://github.com/avolkov-dev/pulumi-mount-issue/blob/master/index.ts#L16 Because now (without

forceDetach

) we are getting Error waiting for Volume (vol-xxx) to detach from Instance: i-xxx error) I believe this is because we trying to detach volume on running instance and getting timeouted

Using python, how can I dynamically get the account id in aws?

☝️ or in general any terraform data source? get caller identity give you the aws account id, but I am not sure how to integrate that with a Pulumi program: https://github.com/pulumi/pulumi-aws/blob/master/sdk/python/pulumi_aws/get_caller_identity.py

Yes -

get_caller_identity

is the answer to how to dynamically get the AWS account id. @broad-finland-69602 what question do you have on how to integrate that? You should just be able to call the function.

thanks @white-balloon-205 and @broad-finland-69602, got it!

from pulumi_aws import get_caller_identity

current = get_caller_identity()

hmm

warning: resource plugin kubernetes is expected to have version >=1.3.1, but has 0.20.3-dev.1550685910+g50fc5e2.dirty; the wrong version may be on your path, or this may be a bug in the plugin

I did a

pulumi plugin rm --all

and

pulumi plugin install

but same occurred, need to rebuild node modules?

Hi everyone, how can prevent I partial creation of a resource? In my case it happens for a database account.

pulumi up

failed for some reason in the first run, and the database account managed to slip through. Subsequent

pulumi up

tried to create that same resource and failed because of name uniqueness.

pulumi refresh

didn't work either How can I handle this?