What is the significance of having this section - spec.template.metadata? It does not seem to be mandatory. However I would like to know where it would be very useful! Otherwise what is the point of repeating all the selectors?
apiVersion: apps/v1
kind: Deployment
metadata:
name: hello
spec:
selector:
matchLabels:
app: hello
tier: backend
track: stable
replicas: 7
template:
metadata:
labels:
app: hello
tier: backend
track: stable
spec:
containers:
- name: hello
image: "gcr.io/google-samples/hello-go-gke:1.0"
ports:
- name: http
containerPort: 80What makes you think it is not required?
If you don't provide the Metadata for a deployment template, it will fail with a message like this:
The Deployment "nginx" is invalid: spec.template.metadata.labels:
Invalid value: map[string]string(nil): `selector` does not match template `lab
els`
Or if the metadata does not match the selector, will fail with a message like this:
The Deployment "nginx" is invalid: spec.template.metadata.labels:
Invalid value: map[string]string{"run":"nginxv1"}: `selector` does not match template `labels`Also, if you do not provide the selector it will error with a message like this:
error validating "STDIN": error validating data: ValidationError(Deployment.spec):
missing required field "selector" in io.k8s.api.apps.v1.DeploymentSpec;
if you choose to ignore these errors, turn validation off with --validate=false
The yaml used is the below:
apiVersion: apps/v1
kind: Deployment
metadata:
labels:
run: nginx
name: nginx
spec:
replicas: 2
selector:
matchLabels:
run: nginx
strategy: {}
template:
metadata:
labels:
run: nginxv1
spec:
containers:
- image: nginx
name: nginxWhen you read the docs, the description for the selector says:
The selector field defines how the Deployment finds which Pods to manage. In this case, you simply select a label that is defined in the Pod template (app: nginx). However, more sophisticated selection rules are possible, as long as the Pod template itself satisfies the rule.
Most objects in Kubernetes have a metadata, it is responsible to store information about the resource like, name, labels, annotations and so on.
When you create a deployment, the template is needed for creation\update of ReplicaSet and PODs, in this case, they need to match the selector, otherwise you would end up with orphan resources around your cluster, and the metadata store the data used to link them.
This was designed this way to make the resources are loosely coupled from each other, if you make simple change to the label of a pod created by this deployment\replicaSet, you will notice that the old POD keep running, but a new one is created, because the old one does not attend the selector rule anymore and ReplicaSet create a new one to keep the number of desired replicas.
That becomes the metadata on the pods that will eventually be created. Usually this is very similar to the metadata on the deployment itself, but it doesn’t have to be.
There are three different objects created with deployment manifest.
In your question, Spec. Template. Metadata is referred to ReplicaSet object. It is recommended to define those parameter to uniquely identifying each or relevant objects for a specific deployment.