Cloud Native Computing Foundation Serverless Practitioners Summit 2019 (San Diego), 24 Nov 2019

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Make Deployments Easy with Project Nebula - Kenaz Kwa, PuppetSDCC

Description

Join us for Kubernetes Forums Seoul, Sydney, Bengaluru and Delhi - learn more at kubecon.io

Don't miss KubeCon + CloudNativeCon 2020 events in Amsterdam March 30 - April 2, Shanghai July 28-30 and Boston November 17-20! Learn more at kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects

Make Deployments Easy with Project Nebula - Kenaz Kwa, PuppetSDCC

Deploying a serverless application requires a wide variety of carefully orchestrated tasks. To get a single change out the door, you need to consider infrastructure provisioning, deployment strategies (Blue/Green, Canary, etc.), secrets management and auditing, logging and monitoring configuration, change management processes and more.

Project Nebula is a new approach to continuously deploying cloud-native applications and infrastructure by composing the tools and workflows your developers already use into simple, repeatable deployment workflows. In this quick overview, we’ll demonstrate how Project Nebula can standardize serverless deployments across your organization, and how it integrates with other tools in the cloud-native ecosystem.

A

Hi everybody, my name is canis and, let's see, make sure this loads. I am a product manager who works at puppet and the first question you might be. Having is well. Why is my talk, five minutes and I'm? Basically the pre-roll ad for your whole day, so you can have a ad free experience, the rest of the time- and you know, give you my pitch and then get off stage.

A

So the second question that you might be asking is what the heck is puppet doing up here and that's a great question, and so over the past couple months, my team has been working on an incubation project that we call project nebula, that's among other things, how to working on a solution for continuously deploying cloud native applications and infrastructure and specifically the user problem that we're trying to solve here. Is you know?

A

How do you orchestrate a serverless deployment as part of a larger deployment, and so, if you have a couple of serverless functions that are being deployed, how then do you sort of take those pieces put them together as part of a larger pipeline, and you know, deploy your application end to end and allow other people in your organization to also you know, leverage those skills and technologies that you people understand how to put together into something that anyone can consume?

A

And so that's the problem that we were looking to solve with with nebula, as I mentioned, a continuous deployment solution that is trying to take the pieces of what puppet has worked on for a long time in the configuration management space for traditional servers and apply some of those similar principles and technologies to a cloud native application deployment, and so, as I mentioned it we're trying to build upon existing tools that you might already have in your ecosystem.

A

And so, if you use terraform for provisioning infrastructure, if you use Hjelm for deploying applications kubernetes, if you use stacker e to do your service functions, you should be able to orchestrate all of those pieces within a within a common workflow.

A

The last third piece here is that all of this product that we're building is built on top of Tek, which is an open-source project for kubernetes native pipelines and there's a whole bunch of other stuff that other folks will come up in here and talk about Tecton, but we're really excited to be able to partner with Tecton and use these technologies to to build this experience on top.

A

So with that I'm gonna switch over and do a demo because there's nothing like failing on a live demo in front of a lot of people and watching them watching you fail and barreling like so in this case, I have a very simple serverless application, a single function that and I'm using stacker you rather than sort of tell you that I know everything about deploying serverless applications, we've partnered with with stacker II, to help you pull this kind of application in an end-to-end way, and- and so in this application.

A

I have a very simple server list function with a single route. It tells me I'll sweep the Wi-Fi is working okay, so hello world, nothing! Nothing too crazy! Here.

A

Okay, give me one second OOP.

A

Better! Okay! Sorry about that.

A

Wi-Fi Wi-Fi, yeah, okay, so as I mentioned, I have a single single function that today I'm deploying using stacker, II and stacker II is a as a server list platform that uses CloudFormation as a way to our packages cloud formation and makes it really easy to deploy server dos applications. I was able to get one running pretty quickly and I have a single server list function here with a single route, as I mentioned that it just gives me a very simple response, so I'm going to modify that so that I make that more venue.

A

Appropriate and I will change that here and then commit this.

A

Fish- and this will now you know, update my my repo and I'm going to now kick off that server list deployment, along with the other pieces that we are required to put that application together. So this is where we wear a project nebula comes in, which is a platform that enables you to orchestrate that server list deployment as part of a broader application deployment.

A

So in this case, I have a stack redeployment as my first step, alongside with that I'm creating a s3 bucket I'm using my cloud formation module or a step, we have support for over 20 different steps of very popular common tools like cloud formation like terraform and appoint a bunch of other things.

A

I am now going to run this workflow with a set of parameters, and what we're doing here is again using Tecton underneath the covers to go, create a pipeline, we sort of model the entire deployment and provide a very simple, yes yeah, Mel type interface, to go on top of that. Well, we know that you might not. You know you spend a lot of all day. You know using animal.

A

We hope that we can provide some visualization tools on top of that, to make it simpler for you to be able to see exactly what's happening in that giant stream of animal. So in this case I mentioned I have creating a website bucket and then deploying Sacre. On top of that, we added an Easter Egg at the very end here that makes this hopefully a little bit more interesting for everybody and I really hope this works. Yes, yes, okay, sweet, okay, so it created the bucket now this daiquiri deployment is happening.

A

We have live logs, so you can see exactly what's happening during this deployment within the stack redeployment. It's now sort of it made the change to to the function.

A

It's now redeploying that entire cloud formation stack, that deploy is the lambda function, among other things, and hopefully, in a couple seconds we will be able to see that the stack was deployed and now, if I go back to that circles function, we see that okay, the response change great so now I deployed that I now deploy the front-end to, and in this case just the s3 bucket I'm, deploying that it's static, HTML file that you'll be able to see and then I notify my team in slack that hey this thing was deployed now everyone can see exactly where to go to go test.

A

This thing out and.

A

When I click boom- yes, so again, this is the the website that we just deployed using the updated. You know copy that we just changed. So, as I mentioned, we are just kicked off the public beta of this last month, and if this is interesting to you, we really appreciate you to be a sign up and give us feedback. You know tell us all the things that we're doing wrong, we'd love to hear them, so we can help. You know you solve some of these problems.

A

For you, our public beta is at public comms, less project nebula and thanks for your time, appreciate.