Looks like it expects program (callable) to be provided for local workspace but that should not be included with work_dir provided.

Has anybody used dictionaries or lists with Output.all...apply? https://pulumi-community.slack.com/archives/CRH5ENVDX/p1632365439151400

How can I enable dynamic partitioning in aws.kinesis.FirehoseDeliveryStream?

Hello community. I have an existing Azure container registry which I imported and I am uising it in my pulumi code. I need to pull a docker image hosted there in order to reference it into an azure app service. Right now, I am creating the azure app service plan and the web app service. Taking as reference this docker azure app service example, they are building the image there I am just wondering whether is possible to pull an existing image as long I got the existing container registry instance. Is that possible? In this case, the

docker.Image

resource they use is not useful since

build

parameter is required. I am seeing this

docker.RemoteImage

resource from docker pulumi api, but not sure how to indicate it that the image should be pulled from my container registry I got previously. The code for getting my container registry is:

container_image = "wmlab"

# GETTING MY CONTAINER REGISTRY
rhdhv_container_registry = azure_native.containerregistry.Registry(
    "rhdhvContainerRegistry",
    admin_user_enabled=True,
    location="westeurope",
    network_rule_set=azure_native.containerregistry.NetworkRuleSetArgs(
        default_action="Allow",
    ),
    registry_name="rhdhvContainerRegistry",
    resource_group_name="genericRG1",
    sku=azure_native.containerregistry.SkuArgs(
        name="Premium",
    ),
    opts=pulumi.ResourceOptions(protect=True))

# OUTPUT THE CREDENTIALS TO GET THEM
acr_credentials = pulumi.Output.all("genericRG1", rhdhv_container_registry.name).apply(
    lambda args: azure_native.containerregistry.list_registry_credentials(
        resource_group_name=args[0],
        registry_name=args[1]
    )
)
admin_username = acr_credentials.username
admin_password = acr_credentials.passwords[0]["value"]

Not sure if perhaps I have to use

azure_native.containerregistry.Registry

(here it is) since it has a

get

method when an instance of a container registry is gotten, such as is shown at the attached picture. The point is I don’t know how to use that get method I am a bit confused about how to pull an image from the existing container registry I got. Or perhaps does it need to be build and pushed from the same project? I would say should be possible to get an existing one. I need this because the build process is taking place in another repository project

Hello everyone!, is there any example repo with complex python unit tests, I am having some trouble trying to mock some resources when creating unit tests for ComponentResources, specially when resolving outputs inside another resource. Or if anyone can share some example, I will appreciate it 🙌

Hey all, started using Pulumi at work for setting up our AWS infrastructure (mainly API's and lambda functions) and it's been a joy so far! We've gotten to a point where we're adding provisioned concurrency to a lambda function following the example here: https://github.com/pulumi/examples/blob/258d3bad0a00020704743e37911c51be63c06bb4/aws-py-serverless-raw/__main__.py#L94 I figured out how to actually publish the function using the publish input argument so that there's a published version for the Provisioned Concurrency Config and noticed that the versions start at 1. The lambda contains a custom python package with a SEMVER version. If I wanted to tag the lambda version to that SEMVER value instead, is there a way to feed in that value along with the

publish

argument? Or is this a use case for creating lambda aliases with that SEMVER value instead?

Hi there - I’m really trying to wrap my head around the

props

argument to the

ComponentResource

constructor: https://www.pulumi.com/docs/reference/pkg/python/pulumi/#the-pulumi-python-resource-model-1 My main questions about it are: • When would I want to pass something to this

props

argument? • What would I pass to that argument?

For example, currently I’m passing a Kubernetes namespace to a “parent”

ComponentResource

that I’ve created. I’m passing it like so:

def __init__(..., namespace: Namespace, ...):
  super().__init__(
    ...,
    props={'namespace': namespace},
    ...
  )

Is that the correct way to use

props

? My thinking was that this would let the Pulumi “engine” know that there is an ordering requirement between the namespace and my

ComponentResource

. Is that how it works? If not, then what are

props

for?

Is there an example exists to Deploy vSphere Virtual Machine from Content Library. It seems like

vsphere.VirtualMachine

exists network_interfaces, and disk params, bug

vsphere.get_content_library_item

does not provide those info.

Does anybody have any Python-based Pulumi stacks with working tests? I want to be able to run unit and integration tests against my stacks. GitHub search came up empty 😞

Hi there - I’m really trying to wrap my head around the 

props

 argument to the 

ComponentResource

 constructor: https://www.pulumi.com/docs/reference/pkg/python/pulumi/#the-pulumi-python-resource-model-1 My main questions about it are: • When would I want to pass something to this 

props

 argument? • What would I pass to that argument? For example, currently I’m passing a Kubernetes namespace to a “parent” 

ComponentResource

 that I’ve created. I’m passing it like so:

def __init__(..., namespace: Namespace, ...):
  super().__init__(
    ...,
    props={'namespace': namespace},
    ...
  )

Is that the correct way to use 

props

? My thinking was that this would let the Pulumi “engine” know that there is an ordering requirement between the namespace and my 

ComponentResource

. Is that how it works? If not, then what are 

props

 for?

I just wanted to share this bit of code that @billowy-army-68599 helped me on for anyone else runs into this issue

Hi, anyone knows a way within an Inline Python program to detect if we are running 'preview' or 'up' ? I have some custom actions within the stack I would ideally only execute during 'up' but not during 'preview' so to NOT alter anything during 'preview'. Thx

hey ya, i'm trying to pass some resource outputs into a docker image build as build args, and keep on getting a json serialisation error as they are passing as outputs rather than the actual value. I've tried using

apply

and

Output.concat

anyone got any pointers?

package_dash_image = Image('package-dash-image',
                               build=DockerBuild(args={'BUCKET': "bucket_name",
                                                       'ACCESS_KEY_ID': package_dash_user_access_key.id,
                                                       'SECRET_ACCESS_KEY': package_dash_user_access_key.secret
                                                       },
                                                 context='source/services/package-dash/',
                                                 ),
                               image_name=package_dash_image_name,
                               registry=package_dash_repository_info,
                               )

re: above thread, something I've been trying to work my head around today is passing outputs into a jinja2 template

hey team, is there any possibility to wrap the python version of the automation API in a async friendly module?

hey ya, has anyone had this error before when running

up

?

aws:lambda:EventSourceMapping (md5-source-event):
    error: connection error: desc = "transport: Error while dialing dial tcp 127.0.0.1:45393: connect: connection refused"

Have it for a couple of other resources that were to be updated as well

Hi Team,

When update the pulumi code for managed es for aws, Pulumi stack deleting and recreating

I just need to update the stack, is there any way? please help

Hello everyone!, quick question, when creating a VPC Route Table, how can I define the the route for local vpc traffic, in the console that target shows up as

local

and in the documentation there is no option for setting that (or at least haven't found one 🙂 )

I believe the local route is created for you automatically.

So its just matter of defining the routes that go outside (through nat or igw), makes sense, will try that, thanks @full-artist-27215!

Yup

I just checked my code... I just create the Route Table, then add a route for an IGW... the local route is there automatically

Awesome, thanks! @full-artist-27215

👍

good luck!

anyone have issues building a container image w/Pulumi? am recently getting an error wrt grpc / protobufs:

#16 59.67 Install protobuf 3.18.1 failed
#16 59.67
#16 59.67 ERRORS:
#16 59.67 add protobuf failed:
#16 59.67 Traceback (most recent call last):
#16 59.67   File "/usr/lib/python3.9/concurrent/futures/thread.py", line 52, in run
#16 59.67     result = self.fn(*self.args, **self.kwargs)
#16 59.67   File "/opt/python/lib/python3.9/site-packages/pdm/installers/synchronizers.py", line 190, in install_candidate
#16 59.67     self.manager.install(can)
#16 59.67   File "/opt/python/lib/python3.9/site-packages/pdm/installers/manager.py", line 38, in install
#16 59.67     installer(candidate.build(), self.environment, candidate.direct_url())
#16 59.67   File "/opt/python/lib/python3.9/site-packages/pdm/installers/installers.py", line 74, in install_wheel
#16 59.67     _install_wheel(
#16 59.67   File "/opt/python/lib/python3.9/site-packages/pdm/installers/installers.py", line 176, in _install_wheel
#16 59.67     for record_elements, stream in source.get_contents():
#16 59.67   File "/opt/python/lib/python3.9/site-packages/installer/sources.py", line 165, in get_contents
#16 59.67     record = record_mapping.pop(item.filename)
#16 59.67 KeyError: 'protobuf.libs/'
#16 59.67

seems to build fine with

3.18.0