5.3. Jsonnet (Optional)
This lab explains how to use jsonnet as manifest format together with Argo CD.
Jsonnet
Jsonnet is a templating language which adds the possibility to programmatically work with the underlying data. It basically is a simple extension of JSON .
Let’s have a look at an example first. The following jsonnet file
{
application1: {
name: "jsonnet-application1",
basepath: "/application/test",
path: self.basepath + "/" + self.name,
},
application2: self.application1 { name: "jsonnet-application2" },
}
will render into:
{
"application1": {
"basepath": "/application/test",
"name": "jsonnet-application1",
"path": "/application/test/jsonnet-application1"
},
"application2": {
"basepath": "/application/test",
"name": "jsonnet-application2",
"path": "/application/test/jsonnet-application2"
}
}
Among many other features, Jsonnet can help to reduce duplications.
Further Docs
Read more about the jsonnet integration in the official documentation .
Task 5.3.1: Deploy the simple-example with jsonnet
Let’s first explore the files in your local repository under jsonnet.
Similar to helm, jsonnet allows us to extract parameters into a separate file. In our example we extracted all values to the params.libsonnet file:
{
containerPort: 5000,
image: "quay.io/acend/example-web-go",
name: "argo-jsonnet-example-<username>",
replicas: 1,
servicePort: 5000,
type: "ClusterIP",
}
And the actual template file, containing the Kubernetes service and deployment definitions, simple-application.jsonnet file:
local params = import 'params.libsonnet';
[
{
"apiVersion": "v1",
"kind": "Service",
"metadata": {
"name": params.name
},
"spec": {
"ports": [
{
"port": params.servicePort,
"protocol": "TCP",
"targetPort": params.containerPort
}
],
"selector": {
"app": params.name
},
"type": params.type
}
},
{
"apiVersion": "apps/v1",
"kind": "Deployment",
"metadata": {
"name": params.name
},
"spec": {
"replicas": params.replicas,
"revisionHistoryLimit": 3,
"selector": {
"matchLabels": {
"app": params.name
},
},
"template": {
"metadata": {
"labels": {
"app": params.name
}
},
"spec": {
"containers": [
{
"image": params.image,
"name": params.name,
"ports": [
{
"containerPort": params.containerPort
}
]
}
]
}
}
}
}
]
Note the first line, where we tell jsonnet where to get params from.
Now replace the username placeholder in the params.libsonnet file with your username.
Commit and push the changes to your repository.
Hint
git add jsonnet
git commit -m "Change Username"
git push
Create the new Argo CD application.
argocd app create argo-jsonnet-$USER --repo https://gitea.training.cluster.acend.ch/$USER/argocd-training-examples.git --path 'jsonnet' --dest-server https://kubernetes.default.svc --dest-namespace $USER
Sync the application.
Hint
To sync (deploy) the resources you can simply click sync in the web UI
or execute the following command:
argocd app sync argo-jsonnet-$USER
And verify whether your jsonnet Application definition has been successfully synced.
Task 5.3.2: Autosync and scale up
Tell the application to sync automatically and to enable self-healing and auto-prune.
Hint
argocd app set argo-jsonnet-$USER --sync-policy automated
argocd app set argo-jsonnet-$USER --self-heal
argocd app set argo-jsonnet-$USER --auto-prune
Now let’s change the replica count of the deployment and scale to 2 pods.
Change the replica param in your params.libsonnet to 2
{
containerPort: 5000,
image: "quay.io/acend/example-web-go",
name: "argo-jsonnet-example-<username>",
replicas: 2,
servicePort: 5000,
type: "ClusterIP",
}
Commit and push the changes to your repository.
Hint
git add jsonnet
git commit -m "Scale Jsonnet to two"
git push
And verify the result in the ArgoCD UI or by using the following command, this might take a little while, depending on how many trainees are currently working on the labs. Hint: hit refresh to speed up the process.
kubectl get pod --namespace $USER --watch
Task 5.3.3: Delete the Applications
Delete the applications after you’ve explored the Argo CD Resources and the managed Kubernetes resources.
Hint
argocd app delete argo-jsonnet-$USER