Control the pod domain

I have pod1 and pod2 in the same namespace.

pod2 is running an HTTP server.

How can I easily get pod2 be seen as pod2.mydomain.com from pod1?

In this way the HTTPS certificate would work with no problem.

It can be achieved through the kubernetes cluster and it is important to use a valid SSL certificate.

By using a Kubernetes service you can create a service of type “ClusterIP” or “NodePort” in the same namespace as the pods and you need to expose the pod2 HTTP server to a consistent IP and port. In this way you can configure your DNS to map pod2.mydomain.com to the IP address of the service.

Or if your cluster supports load balancing you can create a Service of type “LoadBalancer” and you can expose pod2 HTTP server to a public IP.
You can also create an Ingress service that routes traffic to pod 2 on the hostname.
For more information please check this official Document

You can directly hit the POD with it's IP fine how you are thinking however it would better to use the service with POD.

apiVersion: v1
kind: Pod
metadata:
  name: nginx
  labels:
    app.kubernetes.io/name: proxy
spec:
  containers:
  - name: nginx
    image: nginx:stable
    ports:
      - containerPort: 80
        name: http-web-svc

---
apiVersion: v1
kind: Service
metadata:
  name: service-1
spec:
  selector:
    app.kubernetes.io/name: proxy
  ports:
  - name: name-of-service-port
    protocol: TCP
    port: 80
    targetPort: http-web-svc

Service route the traffic to matching labels PODs, so from POD1 you hit the request to service-1 which will forward traffic to POD-1 and response.

https://service-1.<namespace-name>.svc.cluster.local

with service HTTPS will also work the way you asked. Test with the curl command, start one curl pod

kubectl run mycurlpod --image=curlimages/curl -i --tty -- sh

hit curl request to service

curl https://service-1.<namespace-name>.svc.cluster.local

service ref doc : https://kubernetes.io/docs/concepts/services-networking/service/#defining-a-service

Extra :

You can also use the ingress & service mesh which will make little simple for you scenario if you don't want to manage SSL/TLS cert for the app.

Service mesh supports the mTLS auth you can force policy and it would be easy however there will be extra management.

In order not to change SSL certificates, and to use a specific domain, here is the technique I've used successfully:

• install the sudo utility
• Run this instruction in the pod1's Dockerfile, supposing that the user is myuser:

   echo 'myuser   ALL=(ALL:ALL) NOPASSWD:/bin/chmod 666 /etc/hosts' > /etc/sudoers.d/chmod;

• At runtime, from pod1 calculate the IP of pod2 by running:

kubectl get pod pod2 --template {{.status.podIP}}

• Then run:

sudo chmod 666 /etc/hosts

• Then add the following line in /etc/hosts (changing ip by the real IP):

ip pod2.mydomain.com

Maybe this solution is not the most Kubernetes-ish, but it avoids me to deal with certificates.

I consider changing /etc/hosts by far easier than to play with self signed cerficates and all this kind of stuff.