Is there any way to add arbitrary pod labels when using k8s crosswalk

PodBuilder

? From what I can see, it automatically adds an

app

label based off

image

or

name

but there’s no way to add user defined labels

Is it possible to wait for or otherwise query a

k8s.apiextensions.CustomResource

or

k8s.yaml.ConfigFile

to succeed? I see there is an issue tracking to make it type safe. Pulumi applies the resource fine, but I need to then check to make sure it didn’t fail. I see pulumi query, but that doesn’t appear to work with pulumi up.

Is there a way to retreive VM sizes in azure the same way you can query all azure locations (using azure.locations.<locationname>) ? now you have to specify const nodeSize = “Standard_D2_v2” as a string...

So pulumi does not need tfState files like Terraform? Then how does it manage resources?

Is there any progress on this issue? it's blocking all deployments for us. <https://github.com/pulumi/pulumi-aws/issues/821> I can give more detail if this PR isn't explicit enough @microscopic-florist-22719 looked at this 4 days ago. Curious if any insight was gained or even a workaround. <https://pulumi-community.slack.com/archives/C84L4E3N1/p1575580948064900>

Any tips on running

PULUMI_EXPERIMENTAL=true pulumi query

? I am getting

Error: Program run without the Pulumi engine available; re-run using the pulumi CLI

hey it seems that --name doesnt exist with respect to helm install command that pulumi runs

for example this is the error I get

Error: Command failed: helm template /var/folders/2y/61j5tzdd56l8wffgn3ydkf6m0000gp/T/tmp-42616Qai9xo75SzPT/goldilocks --name goldilocks --values /var/folders/2y/61j5tzdd56l8wffgn3ydkf6m0000gp/T/tmp-42616Qai9xo75SzPT/goldilocks/values.yaml --values /var/folders/2y/61j5tzdd56l8wffgn3ydkf6m0000gp/T/tmp-42616JZAjZL6MKzh7.yaml --namespace goldilocks
    Error: unknown flag: --name

--name doesn’t exist inside version 3

You're probably running helm 3, I think only helm 2 is supported right now Looks like the k8s package needs to be updated for

helm3

support

@busy-umbrella-36067 yeah I went down a version

I'm using pulumi on AWS with TypeScript. I have an existing codebuild job that does some stuff. Ultimately, it uploads artifacts to S3. I'd like to launch that job during deployment using resources (domain names) that have been created/updated. Further, I'd like to use the artifacts produced by the job when creating a CloudFront distribution. Any suggestions on how to launch & obtain results from a codebuild job from a pulumi program?

Hey all. I'm trying to work out the best way to carry out a particular use case. I would like to have a single AWS ALB, and then deploy each project independently. Core Infra • create ALB * • create listeners for http and https * per project / services • add a certificate to https listener per service • create target groups per service • create listener rules per service • create Fargate services Any guidance would be much appreciated

I’ve been looking around (in the examples and in the code) for a way to assign a custom url and certificate to an aws api gateway instance. I’m guessing I’m looking in the wrong place, but I’m not sure where the right pace is.

Is it possible that doing

pulumi up --skip-preview

can completely bork the entire state if the up manages to fail?

Hi, is there a way with python and aws, to create two vpc's in different regions? Or do I need to create separate stacks to do this?

hello, I was wondering if it is possible to bridge the TF provider <https://github.com/mumoshu/terraform-provider-helmfile> to Pulumi. Alternatively it might make more sense to take the values from a helmfile.yaml and apply them to pulumi helm, replecating the helmfile functionaliy with pulumi. It wouldn't be the same but it would be good enough to work

What do you think? do the helmfile concepts (https://github.com/roboll/helmfile#highlights) make sense in a Pulumi/Helm setup?

Hey 🙂 !!! Quick question, it’s possible and simple to create an pulumi provider? What is the recommand way?

Can you run stacks sequential? So if I build a VPC in one stack, the second stack adds EC2's and so on. Don't really want one large

___main___.py

file.

i hope everyone accepts the fact eventually AWS Lambda will be stateful and will run webassembly, and we won't need containers, so we won't need container orchestration ad nauseum, so we won't need kubernetes, so we should probably focus on serverless, it's gonna win the war

k8s is a ton of extra work, just to run one process in a container

just my opinion, but it feels like we're wasting hundreds of years collectively, to repeat the same work across many organizations, and containers are a huge waste of money and add a ton of pollution to the environment

For the sake of those of us with alertwords for those technologies, can we please not discuss this in #general ? 🙂

When I get alerts for those words, it usually means someone is in pain and I try to help.

sure, sorry

This may be something dumb on my end, but I have been debugging this for a little while now and am quite frustrated. If someone has suggestions it would be greatly appreciated. When I run pulumi preview locally in a project everything is dandy. However when I run in CI i'm getting

Diagnostics:
  pulumi:pulumi:Stack (xxxxx-staging):
    error: an unhandled error occurred: Program exited with non-zero exit code: -1

I'll thread the verbose logs.

How are people adding custom checks into their pulumi programs. For instance after deploying a k8s resource that has a CRD.

Is there a CLI flag to delete protected resources?