 I'm going to invite Megan from Google, who works on the Kubo Now Container Runtime project team here at the Cloud Foundry Foundation up on stage. And she's going to show you what this new Container Runtime actually is. So Megan, come on up. [?]. [?]. [?]. [?]. [?]. [?]. [?]. [?]. [?]. [?]. [?]. Awesome. I need this box. All right. OK, now here's the part where I prattle and she plugs in. I'm actually really excited to see this demo because one of the questions we get asked is, well, why don't I just run Kubernetes? And whenever you use a platform like Kubernetes, you need to think about, how do you care for the platform itself? And that's what a lot of the focus has been for you and the rest of the project team. Yeah, exactly. Kubo or Cloud Foundry Container Runtime is a project that brings the power of Bosch to Kubernetes. So it does all the things that Bosch does for Cloud Foundry. It can also do for Kubernetes now, too. Super cool. So why don't we just jump into it? What do you say? All right. So one of the things that Bosch can do is help you scale your cluster. Over here, I've tried to deploy 300 pods, which is a lot of pods for no apparent reason. But you can see that I only have 216 available. And that's because Kubernetes limits how many pods can run on a specific worker node. So over here, I have my Bosch manifest. I'll explain this more in a second because it's a lot of YAML. But I'm going to start running the deploy first. So I'm just going to update this to have three worker nodes instead of two. And then over here, I will deploy that manifest. We love our YAML. Yeah. I think so. So you can see Bosch can tell. The only difference that I have is that I now have three instances of the worker instance group instead of two. So I'll say yes. And what it's going to do is it's going to create a new VM for that node. Then it will install the Bosch agent on that. I'm running this on GCP, so it takes about 30 seconds to spin up a VM. And then an additional 20 seconds or so for the agent. Once that's done, it will start running all the Kubernetes processes on it. You can see which jobs it's going to run. Over here, we have Flannel, Docker. The most important one is the KubeLit job. Once that job runs, Kubernetes will see that it's now part of the cluster. And we'll start seeing this number update. But it takes about three minutes total. So one of the other use cases that we couldn't demonstrate, because it takes a little bit more time, was what happens if one of the nodes of the cluster actually disappears? We all know, I mean, I'm sure GCP is amazing, but we all know that public clouds have hardware failures. And that's kind of what you're supposed to design for, right? Yeah, so Bosch can help with that. It has the Resurrector. So if it notices that it doesn't have as many worker nodes as it wants to have, it will spin up a new node as well. It also does that for all the nodes that it's managing, including the Kubernetes masters and XED. Cool. So what are some of the things that the project team's working on right now? Right now, we're working on adding testing for the failure cases, like making sure that everything comes back up correctly, and also rolling updates for Kubernetes versions. I can actually show you over here. This is the Bosch release on GitHub. So for example, if we wanted to update the version, we would see this package update to the next one would probably be 1.8, which came out last week. It did. I was just going to ask how that testing is going. Yeah, we're working on a pipeline for those tests right now. Outstanding. Outstanding. And the intent here is that you're going to be able to, as a user or as an operator of the environment, get these rolling upgrades that have zero downtime as we go from Kubernetes version to Kubernetes version. Yeah, and ideally, it would be zero downtime for your applications, but also for the Kubernetes API. Yep. That's going to be really neat. Yeah. We really need. If we look over here, the way you would update that is here, you define releases in the Bosch manifest. So you would just update, like for this one, this is Kubo at CD, you would update this to v3 or something. This is the Bosch release v2, not at CD v2. We're using at CD v3. All right, how are we doing? It's good. OK, so the deploy finished, and we should start seeing this number tick up within 60 seconds or so. Cool. The anticipation is killing me. I know, because it's just a number. And also, I think all of the demos are really boring, because it's supposed to be boring. Like, it's supposed to do it for you. So all I have to type is deploy, and then it happens. You just wait. Somebody mess with the Wi-Fi? Well, you need to have some fun, right? This is actually not even my local machine, so just kick me off SSH. It's still working on it, so working on it. So one of the things that I know that the. Nodes, though. So yeah, we have the third node now. You can see this one. Let's give it a minute to go. So that's there. Yeah, that works. We want to see the number. Oh, there it goes. And now here goes Kubernetes. It's doing its thing. Now we have one more. It should go up to 300, but we don't have to wait for it. One, because one more is better. Oh, now there's two more. It was happening fast. Oh, there we go. OK. It was faster earlier. We don't have to wait for it to go to 300, but it will. But it's still counting up. So while it counts up, one of the things that one of the newer projects in the cloud native ecosystem is Istio. And I know there's some intention of the project team to do some testing of Istio and to make sure that the way that Kubernetes is being packaged up in the CF container runtime is going to work well with Istio. Yeah, a lot of the work we've been doing is just testing that the Kubernetes cluster works as you expect and testing various workloads on it. Like, for example, Istio, that's one that we were working on very recently. Yeah, so that's super exciting. So that should be great. All right. Well, that was a great demo. We're already up to 280. Yay. It worked seamlessly. So thank you very much. So congratulations. It worked. All of our demos worked today. I know. I thought you jinxed me. So, Megan, thank you so much for coming. Thank you. Appreciate it.