KubeVela
The Open Application Platform based on Kubernetes and OAM.
🚨 Warning: The project is still under heavy development with lots of sharp edges, please don't look inside unless you know what you are doing. Feel free to contact the maintainers in slack channel cncf#kubevela if you are interested in its full story or becoming a vela developer. 🚨
Purpose and goal
- For developers and operators
- KubeVela, as an out-of-box Cloud Native Application Management Platform, provides numerous workloads and operation tooling for application defining, deployment, scaling, traffic, rollout, routing, monitoring, logging, alerting, CI/CD and so on.
- For platform builders
- KubeVela, as a highly extensible PaaS/Serverless Core, provides pluggable capabilities, an elegant way to integrate any workloads and operational capabilities (i.e. traits).
The KubeVela project is fully initialized by the community since day 0 with bootstrapping contributors from 8+ different organizations.
Architecture
For now, KubeVela project has only two components:
- User interface layer
- including:
cli,dashboard,appfile(WIP)- they are all client side tools to provide developer facing abstractions by leveraging CUElang.
- including:
- Vela core
- including:
- OAM Kubernetes runtime to provide application level primitives such as
componentandapplicationetc. - Built-in workload and trait controllers to implement core capabilities such as
webservice,routeandrolloutetc.- This part heavily leverages existing community projects instead of re-inventing the wheels.
- Capability center to register and manage any user provided workloads and traits (WIP).
- OAM Kubernetes runtime to provide application level primitives such as
- including:
The other component we will add in the future is "Infrastructure Layer" by Crossplane to enable KubeVela works smoothly across multiple clouds.
Install
Prerequisites
- Kubernetes cluster running Kubernetes v1.15.0 or greater
- kubectl current context is configured for the target cluster install
kubectl config current-context
Get the Vela CLI
Download the vela binary from the releases page. Unpack the vela binary and add it to $PATH to get started.
sudo mv ./vela /usr/local/bin/vela
Install Vela Core
$ vela install
This command will install vela core controller into your K8s cluster, along with built-in workloads and traits.
Walk through
Note: this is not a getting started guide. Just to let KubeVela developers have some sense of the project.
After vela install you will see available workloads and traits.
$ vela workloads
NAME DEFINITION
backend containerizeds.standard.oam.dev
task jobs.batch.k8s.io
webservice containerizeds.standard.oam.dev
$ vela traits
NAME DEFINITION APPLIES TO
route routes.standard.oam.dev webservice,backend
scale manualscalertraits.core.oam.dev webservice,backend
Create environment
Before working with your application, you should prepare an deploy environment for it (e.g. test, staging, prod etc).
$ vela env init demo --namespace demo --email my@email.com --domain kubevela.io
ENVIROMENT demo CREATED, Namespace: demo, Email: my@email.com.
Vela will create a Kubernetes namespace called demo , with namespace level issuer for certificate generation using the email you provided.
You could check the environment metadata in your local:
$ cat ~/.vela/envs/demo/config.json
{"name":"demo","namespace":"demo","email":"my@email.com","domain":"kubevela.io","issuer":"oam-env-demo"}
Create simple component
Then let's create application, we will use the demo environment.
$ vela comp run mycomp -t webservice --image crccheck/hello-world --port 8000 --app myapp
Creating AppConfig appcomp
SUCCEED
Create micro-services application
Vela supports micro-services application by default thanks to Open Application Model.
$ vela comp run db -t backend --image crccheck/hello-world --app myapp
Creating App myapp
SUCCEED
$ vela comp ls
NAME APP WORKLOAD TRAITS STATUS CREATED-TIME
db myapp backend Deployed 2020-09-18 22:42:04 +0800 CST
mycomp myapp webservice Deployed 2020-09-18 22:42:04 +0800 CST
Under the hood
In Kubernetes, vela creates an OAM application configuration named myapp to manage all related components.
$ kubectl get appconfig -n demo
NAME AGE
myapp 24s
$ kubectl get components -n demo
NAME AGE
mycomp 24s
db 10s
Vela Core is responsible for managing the underlying Kubernetes resources linked with the components and application configuration above.
$ kubectl get deployment -n demo
NAME READY UP-TO-DATE AVAILABLE AGE
mycomp 1/1 1 1 38s
db 1/1 1 1 20s
$ kubectl get svc -n demo
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
mycomp ClusterIP 172.21.4.228 <none> 8080/TCP 49s
Manage operational configurations of the application
Vela leverages OAM trait system to manage operational configurations such as scale, route, canary, autocaleetc in application centric approach.
Let's take route as example.
route
$ vela route mycomp --app myapp
Adding route for app mycomp
Succeeded!
For now you have to check the public address manually (this will be fixed soon so vela route will return visiting URL as result):
$ kubectl get ingress -n demo
NAME HOSTS ADDRESS PORTS AGE
mycomp-trait-5b576c4fc mycomp.kubevela.io 123.57.10.233 80, 443 73s
And after you configure the kubevela.io domain pointing to the public address above.
Your application will be reached by https://mycomp.kubevela.io with mTLS automatically enabled.
Under the hood
Vela will manage the underlying Kubernetes resource which implements the route trait.
$ kubectl get routes.standard.oam.dev -n demo
NAME AGE
mycomp-trait-5b576c4fc 18s
routes.standard.oam.dev is a CRD controller which will manage ingress, domain, certificate etc for your application.
Check status
Check the application:
$ vela app show myapp
About:
Name: myapp
Created at: 2020-09-18 22:42:04.191171 +0800 CST
Updated at: 2020-09-18 22:51:11.128997 +0800 CST
Environment:
Namespace: demo
Components:
Name Type Traits
db backend
mycomp webservice route
Check specific component:
$ vela comp show mycomp
About:
Name: mycomp
WorkloadType: webservice
Application: myapp
Environment:
Namespace: demo
Arguments:
image: crccheck/hello-world
name: mycomp
port: 8000
Traits:
route:
domain: mycomp.kubevela.io
issuer: oam-env-demo
name: route
$ vela comp status mycomp
Showing status of Component mycomp deployed in Environment demo
Component Status:
Name: mycomp Containerized(type) UNKNOWN APIVersion standard.oam.dev/v1alpha1 Kind Containerized workload is unknown for HealthScope
Traits
└─Trait/route
Last Deployment:
Created at: 2020-09-18 22:42:04 +0800 CST
Updated at: 2020-09-18T22:51:11+08:00
Delete application or component
$ vela app ls
myapp
$ vela comp ls
NAME APP WORKLOAD TRAITS STATUS CREATED-TIME
db myapp backend Deployed 2020-09-18 22:42:04 +0800 CST
mycomp myapp webservice route Deployed 2020-09-18 22:42:04 +0800 CST
$ vela comp delete db
Deleting Component 'db' from Application 'db'
$ vela comp ls
NAME APP WORKLOAD TRAITS STATUS CREATED-TIME
mycomp myapp webservice route Deployed 2020-09-18 22:42:04 +0800 CST
$ vela app delete myapp
Deleting Application "myapp"
delete apps succeed myapp from demo
Dashboard
Vela has a simple client side dashboard for you to interact with (note it's still under development). The functionality is equivalent to the vela cli.
$ vela dashboard
Auto-completion
bash
To load completions in your current shell session:
$ source <(vela completion bash)
To load completions for every new session, execute once:
Linux:
$ vela completion bash > /etc/bash_completion.d/vela
MacOS:
$ vela completion bash > /usr/local/etc/bash_completion.d/vela
zsh
To load completions in your current shell session:
$ source <(vela completion zsh)
To load completions for every new session, execute once:
$ vela completion zsh > "${fpath[1]}/_vela"
Clean your environment
$ helm uninstall vela-core -n oam-system
release "vela-core" uninstalled
$ kubectl delete crd workloaddefinitions.core.oam.dev traitdefinitions.core.oam.dev
customresourcedefinition.apiextensions.k8s.io "workloaddefinitions.core.oam.dev" deleted
customresourcedefinition.apiextensions.k8s.io "traitdefinitions.core.oam.dev" deleted
$ rm -r ~/.vela
Contributing
Check out CONTRIBUTING.md to see how to develop with KubeVela.