pulumi-community #general

calm-greece-42329

10/09/2019, 7:37 PM

is there any tips on debugging this stall? i can run a refresh just fine, but

up

hangs at this point and i don’t know what is actually trying to do

calm-greece-42329

10/09/2019, 7:37 PM

this code used to run fine but its been a few weeks since i have run a command in here

calm-greece-42329

10/09/2019, 7:38 PM

or i can blame macos catalina 🙂

breezy-kangaroo-52717

10/09/2019, 11:48 PM

Is the CLI the only interface for pulumi engine? I am hoping to find an API I could use in my web application to call pulumi engine to execute my pulumi programs. How are people implementing such architectures currently if there's no API and CLI is the only way to interact with the pulumi engine?

acceptable-army-69872

10/10/2019, 5:15 PM

I've been playing with the client VPN stuff, and it seems like the features aren't really flushed out. You can create a client vpn, and associate it with a subnet, but you can't assign the security groups, manage the routes, or the authorization. Does that sound right? If so, it's better than nothing, but could be better.

worried-city-86458

10/10/2019, 8:55 PM

Copy code

pulumi:pulumi:Stack (k8s-infra-dev):
    Error: The gRPC binary module was not installed. This may be fixed by running "npm rebuild"
    Original error: Cannot find module '/workspaces/devops-gemini-pulumi/k8s-infra/node_modules/grpc/src/node/extension_binary/node-v72-linux-x64-glibc/grpc_node.node'
    Require stack:
    - /workspaces/devops-gemini-pulumi/k8s-infra/node_modules/grpc/src/grpc_extension.js
    - /workspaces/devops-gemini-pulumi/k8s-infra/node_modules/grpc/src/client_interceptors.js
    - /workspaces/devops-gemini-pulumi/k8s-infra/node_modules/grpc/src/client.js
    - /workspaces/devops-gemini-pulumi/k8s-infra/node_modules/grpc/index.js
    - /workspaces/devops-gemini-pulumi/k8s-infra/node_modules/@pulumi/pulumi/runtime/settings.js
    - /workspaces/devops-gemini-pulumi/k8s-infra/node_modules/@pulumi/pulumi/log/index.js
    - /workspaces/devops-gemini-pulumi/k8s-infra/node_modules/@pulumi/pulumi/cmd/run/run.js
    - /workspaces/devops-gemini-pulumi/k8s-infra/node_modules/@pulumi/pulumi/cmd/run/index.js
        at Object.<anonymous> (/workspaces/devops-gemini-pulumi/k8s-infra/node_modules/grpc/src/grpc_extension.js:43:17)
        at Module._compile (internal/modules/cjs/loader.js:936:30)
        at Object.Module._extensions..js (internal/modules/cjs/loader.js:947:10)
        at Module.load (internal/modules/cjs/loader.js:790:32)
        at Function.Module._load (internal/modules/cjs/loader.js:703:12)
        at Module.require (internal/modules/cjs/loader.js:830:19)
        at require (internal/modules/cjs/helpers.js:68:18)
        at Object.<anonymous> (/workspaces/devops-gemini-pulumi/k8s-infra/node_modules/grpc/src/client_interceptors.js:144:12)
        at Module._compile (internal/modules/cjs/loader.js:936:30)
        at Object.Module._extensions..js (internal/modules/cjs/loader.js:947:10)
 
    error: an unhandled error occurred: Program exited with non-zero exit code: 1

handsome-actor-1155

10/10/2019, 8:58 PM

One thing I keep forgetting to bring up is to ask if anyone has run into a bug(?) where the entry of the VPC exists but the name of vpc is empty in the aws console when creating a VPC from Pulumi? Code:

Copy code

const vpc = new awsx.ec2.Vpc("demo-vpc", {numberOfAvailabilityZones: 3});

worried-city-86458

10/10/2019, 10:18 PM

I'm trying to update the prometheus operator helm chart:

Copy code

pulumi:pulumi:Stack (k8s-infra-dev):
    error: Error: Command failed: helm fetch stable/prometheus-operator --untar --version 6.16.0 --destination /tmp/tmp-333MmxKBHLF631a
    Error: chart "prometheus-operator" matching version "6.16.0" not found in stable index. (try 'helm repo update'). No chart version found for prometheus-operator-6.16.0
 
    Error: chart "prometheus-operator" matching version "6.16.0" not found in stable index. (try 'helm repo update'). No chart version found for prometheus-operator-6.16.0
    Error: Error: Command failed: helm fetch stable/prometheus-operator --untar --version 6.16.0 --destination /tmp/tmp-333MmxKBHLF631a
    Error: chart "prometheus-operator" matching version "6.16.0" not found in stable index. (try 'helm repo update'). No chart version found for prometheus-operator-6.16.0
        at /workspaces/devops-gemini-pulumi/k8s-infra/node_modules/@pulumi/kubernetes/helm/v2/helm.js:112:23
        at OutputImpl.<anonymous> (/workspaces/devops-gemini-pulumi/k8s-infra/node_modules/@pulumi/pulumi/output.js:110:47)
        at Generator.next (<anonymous>)
        at fulfilled (/workspaces/devops-gemini-pulumi/k8s-infra/node_modules/@pulumi/pulumi/output.js:18:58)

early-intern-90238

10/10/2019, 11:10 PM

Why does my GitHub have these status checks but only for one stack? Shouldn't it have a Status check for each stack? I am very confused...

breezy-butcher-78604

10/11/2019, 3:39 AM

hi all, i've run in to an issue trying to create an AWS VPC via AWS Crosswalk. I'm not doing anything complex, just trying to create a VPC with all the defaults:

Copy code

import * as awsx from "@pulumi/awsx";

const vpc = new awsx.ec2.Vpc("my-vpc");

export const vpcId = vpc.id;

When I run

pulumi preview

I get the following error:

Copy code

Previewing update (dev):

     Type                                    Name                Plan       Info
 +   pulumi:pulumi:Stack                     pulumi-vpc-test  create
 +   ├─ awsx:x:ec2:Vpc                       my-vpc              create
 +   │  ├─ awsx:x:ec2:NatGateway             my-vpc-0            create
 +   │  │  └─ aws:ec2:Eip                    my-vpc-0            create
 +   │  ├─ awsx:x:ec2:NatGateway             my-vpc-1            create
 +   │  │  └─ aws:ec2:Eip                    my-vpc-1            create
 +   │  ├─ awsx:x:ec2:InternetGateway        my-vpc              create
 +   │  │  └─ aws:ec2:InternetGateway        my-vpc              create
 +   │  ├─ awsx:x:ec2:Subnet                 my-vpc-public-0     create
 +   │  │  ├─ aws:ec2:RouteTable             my-vpc-public-0     create
 +   │  │  ├─ aws:ec2:Subnet                 my-vpc-public-0     create
 +   │  │  ├─ aws:ec2:Route                  my-vpc-public-0-ig  create
 +   │  │  └─ aws:ec2:RouteTableAssociation  my-vpc-public-0     create
 +   │  ├─ awsx:x:ec2:Subnet                 my-vpc-public-1     create
 +   │  │  ├─ aws:ec2:Subnet                 my-vpc-public-1     create
 +   │  │  ├─ aws:ec2:RouteTable             my-vpc-public-1     create
 +   │  │  ├─ aws:ec2:Route                  my-vpc-public-1-ig  create
 +   │  │  └─ aws:ec2:RouteTableAssociation  my-vpc-public-1     create
 +   │  ├─ aws:ec2:Vpc                       my-vpc              create
 +   │  ├─ awsx:x:ec2:Subnet                 my-vpc-private-1    create
 +   │  │  ├─ aws:ec2:RouteTable             my-vpc-private-1    create
 +   │  │  ├─ aws:ec2:Subnet                 my-vpc-private-1    create
 +   │  │  └─ aws:ec2:RouteTableAssociation  my-vpc-private-1    create
 +   │  └─ awsx:x:ec2:Subnet                 my-vpc-private-0    create
 +   │     ├─ aws:ec2:Subnet                 my-vpc-private-0    create
 +   │     ├─ aws:ec2:RouteTable             my-vpc-private-0    create
 +   │     └─ aws:ec2:RouteTableAssociation  my-vpc-private-0    create
     └─ aws:ec2:NatGateway                   my-vpc-1                       1 error

Diagnostics:
  aws:ec2:NatGateway (my-vpc-1):
    error: aws:ec2/natGateway:NatGateway resource 'my-vpc-1' has a problem: Missing required property 'subnetId'

if I downgrade to v1.2.0 (from 1.3.1) of the pulumi npm module with

npm install @pulumi/pulumi@1.2.0

everything works as expected, the preview completes fine and I can create the VPC without issues. I've had a brief look over the changes between the two versions (https://github.com/pulumi/pulumi-aws/compare/81eb5b...09758c) but there doesn't appear to be any changes to the nodejs code that could explain this. The only thing that seems likely is the change of the terraform provider version used. Has anyone else ran into this issue?

salmon-account-74572

10/11/2019, 5:43 AM

Are there specific words or values that should not be used as the name of a Pulumi project or stack?

cold-coat-35200

10/11/2019, 12:48 PM

hi, is it possible to define a deletion order for kubernetes resources, created from a helm chart? I'm facing a problem with the etcd-operator when using periodic backups. In this case, after the EtcdBackup CR created by pulumi, based on the helm chart, the etcd backup operator adds a finalizer to the EtcdBackup CR (there is no finalizer in the helm chart template). When trying to destroy the stack with pulumi, it's unable to delete the EtcdBackup CR, because the etcd backup operator already deleted(pulumi deletes the resources parallel), so the finalizer never be removed by the operator, therefore kubernetes unable to delete the resource If I destroy the EtcdBackup first with the new

--target

option, then the whole stack, it works fine

salmon-account-74572

10/11/2019, 3:53 PM

When using S3 to store the checkpoint files for Pulumi, it would appear that stack names must be unique across all projects. (I say this because I see stack names listed in

<bucket>/.pulumi/stacks

that aren't differentiated across projects.) Or is the recommendation to use a different S3 bucket for each project?

billions-lock-80282

10/11/2019, 3:54 PM

Hi, I have an issue when running pulumi preview, the stack just sits at 'create' and doesn't move past this. Is there any way of finding out whats happening behind the scenes?

busy-umbrella-36067

10/11/2019, 4:05 PM

Helm chart previews are now fast with the new Pulumi version 👏

🚀 1

🍺 1

early-intern-90238

10/11/2019, 4:06 PM

I actually can't stand Helm, I only use it if I don't have a choice...Way too unreliable...and the authors of the Helms don't seem to give two craps about breaking changes between revisions...

👆 1

early-intern-90238

10/11/2019, 4:07 PM

But having the support is obviously a nice thing to have when they are needed.

busy-umbrella-36067

10/11/2019, 4:08 PM

@early-intern-90238 Same boat here, but I get to avoid managing manifests for

external-dns

kiam

cert-manager

datadog

traefik

, etc...

early-intern-90238

10/11/2019, 4:12 PM

traefik is a good example of unavoidable

damp-pillow-67781

10/11/2019, 5:26 PM

Hi, I've run into the issue that pulumi up is hanging forever until I ctrl+c and stop it. It's on my newly provisioned centos7 machine. Any idea on why that's happenning? It runs fine on my mac.

Copy code

➜  eks-cluster git:(develop) ✗ pulumi up --logtostderr -v=7 --debug 2> out.txt
Previewing update (tableau/online-sandbox-yding):

     Type                 Name                              Plan       Info
 +   pulumi:pulumi:Stack  eks-cluster-online-sandbox-yding  create...  debug: RegisterResource RPC finished: resource:eks-cluster-online-sandbox-yd

cool-egg-852

10/11/2019, 7:53 PM

Is it on purpose that

ResourceRequirements

has the type as

pulumi.Input<object>

instead of having a named object with

cpu

and

memory

adamant-dress-73325

10/11/2019, 9:30 PM

Hi, my pulumi program is stuck creating a new stack? I am running pulumi up on a new stack. Is there something wrong with app.pulumi.com?

early-intern-90238

10/11/2019, 11:17 PM

I wish I could suppress output on a certain very small subset of exports so they don't show up in my GitHub PR's but the rest I want to see...

bored-furniture-11856

10/12/2019, 6:17 AM

Hi folks. I try to use a different S3 provider (Digital Ocean, Minio) than Amazon AWS as state store for Pulumi, but fail to configure the S3 host/URL. With the debug log enabled, you can see that Pulumi is still contacting a host of Amazon S3. Even my quick look at the implementation seems to confirm that currently only AWS S3 is supported as S3 provider? Is that supposed to work? Did I miss a thing?

early-intern-90238

10/12/2019, 3:53 PM

Ok I have a really stupid question, how can I extract an export from the stack that was previously exported in a prior run?

flaky-continent-74538

10/14/2019, 3:36 AM

Is pulumi down? I'm getting tons of weird errors when following the quickstart guide

incalculable-diamond-5088

10/14/2019, 3:29 PM

Getting the following error when running pulumi up:

Copy code

TypeError: Class extends value undefined is not a constructor or null
        at Object.<anonymous> (/Users/igor/work/some-project/infra/pulumi/node_modules/@pulumi/cloudwatch/eventRuleMixins.ts:70:56)
        at Module._compile (internal/modules/cjs/loader.js:868:30)
        at Object.Module._extensions..js (internal/modules/cjs/loader.js:879:10)
        at Module.load (internal/modules/cjs/loader.js:731:32)
        at Function.Module._load (internal/modules/cjs/loader.js:644:12)
        at Module.require (internal/modules/cjs/loader.js:771:19)
        at require (internal/modules/cjs/helpers.js:68:18)
        at Object.<anonymous> (/Users/igor/work/some-project/infra/pulumi/node_modules/@pulumi/cloudwatch/cloudwatchMixins.ts:16:1)
        at Module._compile (internal/modules/cjs/loader.js:868:30)
        at Object.Module._extensions..js (internal/modules/cjs/loader.js:879:10)

acceptable-army-69872

10/14/2019, 4:05 PM

Prefix: We aren't doing branching right; I'm trying to see what would happen if I created an RDS instance in a branch my co-worker is using. So I do a

pulumi up

which tells me it's going to update like 500 resources (obviously source has drifted), and create my resource. Which is almost good enough to do a PR, but like, it'd be cool if I can

-target

a resource that doesn't exist yet and try to create it. Can I do that? If so, how do I predict a URN.

loud-computer-30956

10/14/2019, 4:12 PM

Hi all. I'm starting from scratch to make a simple Django app that will be hosted on ElasticBeanstalk. So far so good, if I have the Django app zipped in my Pulumi folder. but going forward, how should I setup my continuous delivery pipelines? Should pulumi up be called or should I just use the AWS cli directly? I guess ideally I use something like GitHub Packages or artifactory to host the zip before invoking pulumi/awscli?

calm-greece-42329

10/14/2019, 4:31 PM

is anyone out there using pulumi in a monorepo? i saw an open issue related to the node_modules, but im curious how people manage this or building a dependency tree to know which stacks need updating. i see some bazel rules for terraform but im also not 100% sold on using a build tool for building infrastructure