Is the ingress class or specifying the ingress controller directly in the application more commonly used now?

The ingress class (ingressClassName) is used more frequently.

This allows:

• To separate ingress configuration from application configuration.
• Easier management of different ingress controllers within a cluster.
• Flexibly apply different ingress controller configurations for different ingresses (e.g., different SSL certificates, different load balancing types).

Specifying the ingress controller via annotations (e.g., kubernetes.io/ingress.class) is considered an outdated approach in most scenarios, especially with the introduction of ingressClassName in the networking.k8s.io/v1 API.

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: minimal-ingress
spec:
  ingressClassName: nginx # prefer this approach
  rules:
  - http:
      paths:
      - path: /test
        pathType: Prefix
        backend:
          service:
            name: test-service
            port:
              number: 80

apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: older-ingress
  annotations:
    kubernetes.io/ingress.class: nginx # older approach, still supported by some controllers
spec:
  rules:
  - http:
      paths:
      - path: /test
        pathType: Prefix
        backend:
          service:
            name: test-service
            port:
              number: 80

Comparison table:

Aspect	ingressClassName	Annotation kubernetes.io/ingress.class
Status	Recommended	Deprecated (in most cases)
Flexibility	High	Medium
Readability	Better	Worse
API	Part of API specification	Annotation