pulumi-community #general

Docs

Join the conversation Join Slack

Channels

general

cuddly-eye-68174

12/22/2018, 9:51 AM

I destroyed my Cluster, and the configmap is still tehre, and it breaks the stack

cuddly-eye-68174

12/22/2018, 9:52 AM

I think explicit dependency is needed for K8s resources with its cluster

cuddly-eye-68174

12/22/2018, 9:53 AM

so the cluster should not be deleted before k8s resources

faint-motherboard-95438

12/22/2018, 4:05 PM

Does anyone knows what this is about ?

E1222 16:44:42.989623   49002 streamwatcher.go:109] Unable to decode an event from the watch stream: http2: server sent GOAWAY and closed the connection; LastStreamID=47, ErrCode=NO_ERROR, debug=""

cuddly-eye-68174

12/22/2018, 4:11 PM

cuddly-eye-68174

12/22/2018, 4:11 PM

Guys How can I give subnet to a launch configuration? In Pulumi-AWS, Python

cuddly-eye-68174

12/22/2018, 4:18 PM

I tried to find it here: https://github.com/pulumi/pulumi-aws/blob/master/sdk/python/pulumi_aws/ec2/launch_configuration.py

cuddly-eye-68174

12/22/2018, 4:20 PM

I cant see that, but maybe you have another solution for this i would like to achieve. So my question is. What is the best way to force ASG to launch instances in a specific VPC, in a specific subnet or subnets in a specific availability zone?

cuddly-eye-68174

12/22/2018, 4:50 PM

Guys, my college helped me out again: so you have solution for this here:

cuddly-eye-68174

12/22/2018, 4:50 PM

https://github.com/pulumi/pulumi-aws/blob/master/sdk/python/pulumi_aws/autoscaling/group.py#L82

cuddly-eye-68174

12/22/2018, 4:51 PM

But logically it should put into the launch config If you would ask me 🙂

cuddly-eye-68174

12/22/2018, 5:17 PM

But I think the whole engine is based on Terraform....so maybe this is not the right place to suggest this and bother you 🙂 Sorry...

shy-finland-77998

12/22/2018, 8:34 PM

It doesn’t appear pulumi is recognizing a custom tsconfig.json file and I’m getting errors such as

error TS2304: Cannot find name 'APIEndpoint'

For context, APIEndpoint is an interface type in a folder in the project When I run the project outside of pulumi (using

ts watch

, etc.) these types get picked up and there’s no issue. So best I can tell when running pulumi, it doesn’t know about these files for whatever reason. I’m troubleshooting it now but any advice in the meantime?

shy-finland-77998

12/22/2018, 9:09 PM

here’s another one I’m running in to. This looks like it happens after compilation:

error: Error serializing '(ev, ctx, cb) => { let body; ...': api.js(189,21)

    '(ev, ctx, cb) => { let body; ...': api.js(189,21): captured
      variable 'handlers' which indirectly referenced
        function '<anonymous>': organizations.ts(5,14): which could not be serialized because
          arrow function captured 'this'. Assign 'this' to another name outside function and capture that.

    Function code:
      (req, res) => tslib_1.__awaiter(this, void 0, void 0, function* () {
          try {
              const results = yield organization_1.getOrganizations();
              res.status(200).json({
                  results: results

Looking at the compiled JS, it looks like TS is doing this:

exports.all = (req, res) => tslib_1.__awaiter(this, void 0, void 0, function* () {

And Pulumi isn’t liking the

this

that it adds in there.

cuddly-eye-68174

12/23/2018, 10:02 AM

Guys, If i would like to turn off the random string suffix on AWS resources. Of course taking the risk there is another resource with the same name in a different stack. How can I do this if it is possible? Lets say I would like to make a role KubeAdmins not KubeAdmin-sd4575s 🙂

square-machine-57767

12/23/2018, 9:08 PM

Hi all! We are looking into using Pulumi in airgapped environments, and also for environments which consider knowledge of their infrastructure sensitive and as such using the SaaS state / stack management offering is not an option. I'm aware of the filesystem backend option - are there any plans (and ETA) for supporting S3 and GCS backends?

shy-finland-77998

12/23/2018, 10:59 PM

Got another one playing around with the container example:

Error: invocation of aws:ec2/getVpc:getVpc returned an error: invoking aws:ec2/getVpc:getVpc: no matching VPC found

This is when trying to use a container for fargate

shy-finland-77998

12/23/2018, 10:59 PM

i double checked the perms on this user where I’m trying to create the fargate instance and it can create a VPC. I assume its failing somewhere in the chain.

tall-librarian-49374

12/24/2018, 8:53 AM

Since I updated 0.16.1 -> 0.16.8 I get some weird symbols printed in my powershell:

←[38;5;13m←[1mUpdating (mystackname):←[0m

tall-librarian-49374

12/24/2018, 8:53 AM

Also feels like my Azure deployments got much slower? Might be unrelated, but feels like it is.

tall-librarian-49374

12/24/2018, 4:27 PM

If I deleted a resource manually, shouldn't

pulumi refresh

delete it from the stack? Instead of failing with

Resource was not found

polite-helmet-87481

12/24/2018, 4:53 PM

root@dev:~/test/examples/aws-js-webserver# npm install
npm WARN package.json webserver@0.1.0 No repository field.
npm http GET <https://registry.npmjs.org/pulumi/pulumi>
npm http GET <https://registry.npmjs.org/pulumi/aws>
npm http GET <https://registry.npmjs.org/pulumi/pulumi>
npm http GET <https://registry.npmjs.org/pulumi/aws>
npm http GET <https://registry.npmjs.org/pulumi/aws>
npm http GET <https://registry.npmjs.org/pulumi/pulumi>
npm ERR! Error: failed to fetch from registry: pulumi/aws
npm ERR!     at RegClient.<anonymous> (/usr/share/npm/node_modules/npm-registry-client/lib/get.js:134:18)
npm ERR!     at cb (/usr/share/npm/node_modules/npm-registry-client/lib/request.js:27:9)
npm ERR!     at RegClient.<anonymous> (/usr/share/npm/node_modules/npm-registry-client/lib/request.js:148:10)
npm ERR!     at cb (/usr/share/npm/node_modules/npm-registry-client/lib/request.js:158:9)
npm ERR!     at RegClient.<anonymous> (/usr/share/npm/node_modules/npm-registry-client/lib/request.js:223:20)
npm ERR!     at self.callback (/usr/lib/nodejs/request/index.js:148:22)
npm ERR!     at Request.EventEmitter.emit (events.js:117:20)
npm ERR!     at ClientRequest.self.clientErrorHandler (/usr/lib/nodejs/request/index.js:258:10)
npm ERR!     at ClientRequest.EventEmitter.emit (events.js:95:17)
npm ERR!     at CleartextStream.socketErrorListener (http.js:1561:9)
npm ERR! If you need help, you may report this log at:
npm ERR!     <http://github.com/isaacs/npm/issues>
npm ERR! or email it to:
npm ERR!     <npm-@googlegroups.com>

npm ERR! System Linux 3.13.0-163-generic
npm ERR! command "/usr/bin/nodejs" "/usr/bin/npm" "install"
npm ERR! cwd /home/user/test/examples/aws-js-webserver
npm ERR! node -v v0.10.25
npm ERR! npm -v 1.3.10
npm ERR!
npm ERR! Additional logging details can be found in:
npm ERR!     /home/user/test/examples/aws-js-webserver/npm-debug.log
npm ERR! not ok code 0

Also did a little digging

root@dev:~/test/examples/aws-js-webserver# curl -s <https://registry.npmjs.org/pulumi/pulumi>
{"code":"MethodNotAllowedError","message":"GET is not allowed"}
root@dev:~/test/examples/aws-js-webserver# curl -i <https://registry.npmjs.org/pulumi/pulumi>
HTTP/1.1 405 Method Not Allowed
Date: Mon, 24 Dec 2018 16:51:30 GMT
Content-Type: application/json
Content-Length: 63
Connection: keep-alive
Set-Cookie: __cfduid=d745c4d85092196555247662fecaeb3b21545670290; expires=Tue, 24-Dec-19 16:51:30 GMT; path=/; domain=.<http://registry.npmjs.org|registry.npmjs.org>; HttpOnly
Allow: PUT
CF-RAY: 48e492b3a9297967-SEA
Expect-CT: max-age=604800, report-uri="<https://report-uri.cloudflare.com/cdn-cgi/beacon/expect-ct>"
Server: cloudflare
root@dev:~/test/examples/aws-js-webserver# curl -is <https://registry.npmjs.org/pulumi/aws>
HTTP/1.1 405 Method Not Allowed
Date: Mon, 24 Dec 2018 16:53:01 GMT
Content-Type: application/json
Content-Length: 63
Connection: keep-alive
Set-Cookie: __cfduid=d3bfe6cf107fba72158cdd2bb7c000f5d1545670381; expires=Tue, 24-Dec-19 16:53:01 GMT; path=/; domain=.<http://registry.npmjs.org|registry.npmjs.org>; HttpOnly
Allow: PUT
CF-RAY: 48e494e9cf0c79cd-SEA
Expect-CT: max-age=604800, report-uri="<https://report-uri.cloudflare.com/cdn-cgi/beacon/expect-ct>"
Server: cloudflare

{"code":"MethodNotAllowedError","message":"GET is not allowed"}

polite-helmet-87481

12/24/2018, 5:03 PM

`update`: solved this by purging and reinstalling

nodejs

and

npm

👍 1

polite-helmet-87481

12/24/2018, 5:03 PM

$ sudo apt-get purge nodejs npm
$ curl -sL <https://deb.nodesource.com/setup> | sudo bash -
$ sudo apt-get install -y nodejs

faint-motherboard-95438

12/26/2018, 11:11 AM

guys, pulumi states that the

0.16.9

version is available but it seems

brew

does not know about that. Do you plan to release it soon on this package manager ?

busy-umbrella-36067

12/26/2018, 8:44 PM

damn

0.16.9

is quite the performance improvement. definitely noticeable once you hit more than 100 resources

👍 3

👏 1

orange-tailor-85423

12/27/2018, 4:19 AM

Welcome — AWS Cloud Development Kit https://awslabs.github.io/aws-cdk/ Thoughts ?

broad-mouse-60437

12/27/2018, 7:38 AM

Guys, does pulumi provides the inter operable language support ?

broad-mouse-60437

12/27/2018, 7:39 AM

Like if I want to use my Python component in my JS script, does it support this ?

faint-motherboard-95438

12/27/2018, 5:55 PM

<http://app.pulumi.com|app.pulumi.com>

is nearly unreachable (very slow and basically unresponsive), have you any issue since 10min ?

Title

faint-motherboard-95438

12/27/2018, 5:55 PM

<http://app.pulumi.com|app.pulumi.com>

is nearly unreachable (very slow and basically unresponsive), have you any issue since 10min ?

white-balloon-205

12/27/2018, 5:58 PM

Not seeing any issues here - and we haven’t seen any alerts from our external monitoring. Are you seeing issues on the site itself? Or via the CLI? Or both? If the site, could you see if any particular network requests are slow for you? (In browser dev tools networking tab)?

faint-motherboard-95438

12/27/2018, 5:59 PM

@white-balloon-205 unable to have

<https://app.pulumi.com>

to finish loading on a browser. a

pulumi stack ls

takes >20sec to complete I have no issue on any other services, my connection is fine, I tested without vpn.

white-balloon-205

12/27/2018, 6:02 PM

Curious - where geographically are you located? If you are able to open browser dev tools networking tab, can you tell if slow requests are to just api.pulumi.com or to other resources?

faint-motherboard-95438

12/27/2018, 6:05 PM

And still loading, nothing displayed yet

I’m located in France (Paris area)

pulumi stack ls

is up to 1min now

white-balloon-205

12/27/2018, 6:09 PM

Looking...

faint-motherboard-95438

12/27/2018, 6:47 PM

kind of clueless. It feels like a load issue on your side. I can reproduce on different computers and different networks.

<https://app.pulumi.com>

takes on average 100sec to complete with all dependencies. and still a minute for a

pulumi stack ls

white-balloon-205

12/27/2018, 6:49 PM

We haven't seen any requests on app.pulumi.com or api.pulumi.com take >100ms to serve in the last 3 hours. I tried an EC2 VM in Paris region and I don't see any slowness running

pulumi

CLI or loading pages from

<http://app.pulumi.com|app.pulumi.com>

faint-motherboard-95438

12/27/2018, 6:56 PM

As I said, i can reproduce on different networks and different computers with totally different configurations so I definitely eliminate an issue from my main access and terminal. I don’t know why you can’t see nor reproduce it.

white-balloon-205

12/27/2018, 7:00 PM

Yeah - still investigating - but truly nothing showing any signs of slowness on our end yet.

Tried an OVH VM in Strasbourg as well.

pulumi stack ls

returns 1450 records for me in 2s.

View count: 1