Can I rename a stack?

Would it be possible to use Pulumi for "Playbook" style, one-off tasks? Like if I wanted to script up getting a Kubernetes deployment, scaling to 0, pointing the service to a maintenance page; would that be a sensible use-case?

What does this error mean?

Previewing destroy of stack 'ComputeSoftware/command-processor-prod'
error: could not find plugin for provider 'urn:pulumi:command-processor-prod::command-processor::pulumi:providers:aws::default'

Great post from the folks at Wallaroo on using Pulumi to spin up on-demand analytics clusters. https://blog.wallaroolabs.com/2018/10/spinning-up-a-wallaroo-cluster-is-easy/

If you just need to edit the search, if you have the URL for it, you can just tack on &e=T to the end of the URL

What is Pulumi’s Team pricing headed towards? We’re thinking of adopting Pulumi, but it’s not clear what we get for free or what we’ll wind up paying on the Team plan.

checking in on https://github.com/pulumi/pulumi-azure/issues/134 arurerm provider update - is this something we can ever do on our own by running the tfgen tools? or will we always have to wait for a pulumi update?

@busy-salesclerk-80625 Thank you so much... does not sound like we need that hahaha

I added

protocol: "http"

to my

cloudAws.Service

ports

and now I am getting a 502 when I visit the URL. When I remove the

protocol

from the

ports

it works as expected. Code:

let appService = new cloud.Service("basic-web", {
    containers: {
        basic_web: {
            build: "../app",
            memory: 256,
            ports: [{port: 80, targetPort: 8080, external: true, protocol: "http"}]
        }
    },
    replicas: 3
});

export let url = appService.defaultEndpoint.apply(e => `http://${e.hostname}`);

Any idea why this would happen?

ECS is constantly restarting containers due to a health check failure. Current guess is the ALB Pulumi creates does not have the correct health check set up.

I have a feeling I need to specify

healthCheckGracePeriodSeconds

for the in the service definition. Is there a way to do this?

The ServiceArgs property back has a field you can set: https://pulumi.io/reference/pkg/nodejs/@pulumi/aws/ecs/#property-healthcheckgraceperiodseconds-1

But I don't think it's exposed at the cloud layer.

Yeah I'm using

cloudAws.Service

.

anybody heading to KubeCon? Would love to talk Pulumi

the hallway track is often better than sessions

any way of using the nice

cloud.Service

abstraction without spinning up a load balancer and making the service public?

I can’t find the changes for 0.16

the changelog?

Yeah

We’ll be publishing changelog shortly! (As well as flipping the default install over to 0.16).

Thanks!

brew upgrade

just gave me 0.16 🙂

aws kinesis data analytics support seems to be missing though I see the terraform provider seems to support it, any idea when that is likely to get added?

We're excited to be working with GitHub on GitHub Actions. This post from @big-piano-35669 covers how to do end-to-end continuous delivery with Pulumi and Actions (example is RoR App > Container > Kubernetes, all from code commits). We'll be featured in this morning's keynote at GH Universe too. https://blog.pulumi.com/continuous-delivery-to-any-cloud-using-github-actions-and-pulumi

Pulumi

0.16

is available now - see https://pulumi-community.slack.com/archives/CB36DSVSA/p1539800854000100 for more details!

@creamy-potato-29402 it seems that when a Deployment depends on a Secret, and the secret is changed, pulumi doesn’t restart / recreate the deployment (unlike for ConfigMaps, where it does so). Shouldn’t this be the same behaviour between configmap and secret?

any way to fix a failed update?

error: [409] Conflict: Another update is currently in progress.