Skip to main content
Version: v0.12

Deploy Application

This guide shows you how to use Kusion CLIs to complete the deployment of an application running in Kubernetes. We call the abstraction of application operation and maintenance configuration as AppConfiguration, and its instance as Application. It is essentially a configuration model that describes an application. The complete definition can be seen here.

In production, the application generally includes minimally several k8s resources:

  • Namespace
  • Deployment
  • Service
tip

This guide requires you to have a basic understanding of Kubernetes. If you are not familiar with the relevant concepts, please refer to the links below:

Prerequisites​

Before we start, we need to complete the following steps:

1、Install Kusion

We recommend using HomeBrew(Mac), Scoop(Windows), or an installation shell script to download and install Kusion. See Download and Install for more details.

2、Running Kubernetes cluster

There must be a running and accessible Kubernetes cluster and a kubectl command line tool. If you don't have a cluster yet, you can use Minikube to start one of your own.

Initializing​

This guide is to deploy an app using Kusion, relying on the Kusion CLI and an existing Kubernetes cluster.

Initializing workspace configuration​

In version 0.10.0, we have introduced the new concept of workspaces, which is a logical layer whose configurations represent an opinionated set of defaults, often appointed by the platform team. In most cases workspaces are represented with an "environment" in traditional SDLC terms. These workspaces provide a means to separate the concerns between the application developers who wish to focus on business logic, and a group of platform engineers who wish to standardize the applications on the platform.

Driven by the discipline of Platform Engineering, management of the workspaces, including create/updating/deleting workspaces and their configurations should be done by dedicated platform engineers in a large software organizations to facilitate a more mature and scalable collaboration pattern.

tip

More on the collaboration pattern can be found in the design doc.

However, if that does NOT apply to your scenario, e.g. if you work in a smaller org without platform engineers or if you are an individual developer, we wish Kusion can still be a value tool to have when delivering an application. In this guide, we are NOT distinctively highlighting the different roles or what the best practices entails (the design doc above has all that) but rather the steps needed to get Kusion tool to work.

As of version 0.11.0, workspace configurations in Kusion can not only be managed on the local filesystem in the form of YAML files, but the remotely-managed workspaces have been supported as well.

To initialize the workspace configuration:

~/playground$ touch ~/dev.yaml
~/playground$ kusion workspace create dev -f ~/dev.yaml
create workspace dev successfully

To verify the workspace has been created properly:

~/playground$ kusion workspace list
- default
- dev
~/playground$ kusion workspace show dev
{}

Note that show command tells us the workspace configuration is currently empty, which is expected because we created the dev workspace with an empty YAML file. An empty workspace configuration will suffice in some cases, where no platform configurations are needed.

Kusion by default uses the default workspace, thus we need to switch to the dev workspace we have just created.

~/playground$ kusion workspace switch dev

We will progressively add more workspace configurations throughout this user guide.

Initializing application configuration​

Now that workspaces are properly initialized, we can begin by initializing the application configuration:

# Create a new directory and navigate into it. 
mkdir simple-service && cd simple-service

# Initialize the demo project with the name of the current directory.
kusion init

The directory structure is as follows:

simple-service/
.
β”œβ”€β”€ dev
β”‚Β Β  β”œβ”€β”€ kcl.mod
β”‚Β Β  β”œβ”€β”€ main.k
β”‚Β Β  └── stack.yaml
└── project.yaml

2 directories, 4 files

The project directory has the following files that are automatically generated:

  • project.yaml represents project-level configurations.
  • dev directory stores the customized stack configuration:
    • dev/main.k stores configurations in the dev stack.
    • dev/stack.yaml stores stack-level configurations.
    • dev/kcl.mod stores stack-level dependencies.

In general, the .k files are the KCL source code that represents the application configuration, and the .yaml is the static configuration file that describes behavior at the project or stack level.

info

See Project and Stack for more details about Project and Stack.

The kusion init command will create a demo quickstart application, we may update the dev/kcl.mod and dev/main.k later.

kcl.mod​

There should be a kcl.mod file generated automatically under the project directory. The kcl.mod file describes the dependency for the current project or stack. By default, it should contain a reference to the official kam repository which holds the Kusion AppConfiguration and related workload model definitions that fits best practices. You can also create your own models library and reference that.

You can change the package name in kcl.mod to simple-service:

dev/kcl.mod

[package]
name = "simple-service"
version = "0.1.0"

[dependencies]
kam = { git = "https://github.com/KusionStack/kam.git", tag = "0.2.0" }
service = { oci = "oci://ghcr.io/kusionstack/service", tag = "0.1.0" }
network = { oci = "oci://ghcr.io/kusionstack/network", tag = "0.2.0" }

[profile]
entries = ["main.k"]

main.k​

The configuration file main.k, usually written by the application developers, declare customized configurations for a specific stack, including an Application instance of AppConfiguration model.

You can update the main.k as follows:

import kam.v1.app_configuration as ac
import service
import service.container as c
import network as n

"helloworld": ac.AppConfiguration {
workload: service.Service {
containers: {
"helloworld": c.Container {
image = "gcr.io/google-samples/gb-frontend:v4"
}
}
replicas: 2
}
accessories: {
"network": n.Network {
ports: [
n.Port {
port: 80
}
]
}
}
}

Previewing​

At this point, the project has been completely initialized. The configuration is written in KCL, not JSON/YAML which Kubernetes recognizes, so it needs to be built to get the final output. And we can use the kusion preview cmd to preview the Kubernetes resources intended to deliver.

Enter stack dir simple-service/dev and preview:

cd simple-service/dev && kusion preview
tip

For instructions on the kusion command line tool, execute kusion -h, or refer to the tool's online documentation.

Applying​

Preview is now completed. We can apply the configuration as the next step. In the output from kusion preview, you can see 3 resources:

  • a Namespace named simple-service
  • a Deployment named simple-service-dev-helloworld in the simple-service namespace
  • a Service named simple-service-dev-helloworld-private in the simple-service namespace

Execute command:

kusion apply

The output is similar to:

 βœ”οΈŽ  Generating Spec in the Stack dev...                                                                                                                                                                                                     
Stack: dev ID Action
* β”œβ”€ v1:Namespace:simple-service Create
* β”œβ”€ v1:Service:simple-service:simple-service-dev-helloworld-private Create
* └─ apps/v1:Deployment:simple-service:simple-service-dev-helloworld Create


? Do you want to apply these diffs? yes
Start applying diffs ...
SUCCESS Create v1:Namespace:simple-service success
SUCCESS Create v1:Service:simple-service:simple-service-dev-helloworld-private success
SUCCESS Create apps/v1:Deployment:simple-service:simple-service-dev-helloworld success
Create apps/v1:Deployment:simple-service:simple-service-dev-helloworld success [3/3] β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ 100% | 0s
Apply complete! Resources: 3 created, 0 updated, 0 deleted.

After the configuration applying successfully, you can use the kubectl to check the actual status of these resources.

1、 Check Namespace

kubectl get ns

The output is similar to:

NAME                   STATUS   AGE
default Active 117d
simple-service Active 38s
kube-system Active 117d
...

2、Check Deployment

kubectl get deploy -n simple-service

The output is similar to:

NAME                            READY   UP-TO-DATE   AVAILABLE   AGE
simple-service-dev-helloworld 1/1 1 1 59s

3、Check Service

kubectl get svc -n simple-service

The output is similar to:

NAME                                    TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)   AGE
simple-service-dev-helloworld-private ClusterIP 10.98.89.104 <none> 80/TCP 79s

4、Validate app

Using the kubectl tool, forward native port 30000 to the service port 80.

kubectl port-forward svc/simple-service-dev-helloworld-private -n simple-service 30000:80

Open browser and visit http://127.0.0.1:30000:

app-preview