Auto-Devops TLS Hosts Invalid value

Summary

Without any changes to our pipeline setup (still using Auto-devops with the same configurations), the pipeline now fails with the following error message:

+ helm upgrade --install --wait --set ...
UPGRADE FAILED
ROLLING BACK
Error: Ingress.extensions "production-auto-deploy" is invalid: spec.tls[1].hosts: Invalid value: "": a DNS-1123 subdomain must consist of lower case alphanumeric characters, '-' or '.', and must start and end with an alphanumeric character (e.g. 'example.com', regex used for validation is '[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*')
Error: UPGRADE FAILED: Ingress.extensions "production-auto-deploy" is invalid: spec.tls[1].hosts: Invalid value: "": a DNS-1123 subdomain must consist of lower case alphanumeric characters, '-' or '.', and must start and end with an alphanumeric character (e.g. 'example.com', regex used for validation is '[a-z0-9]([-a-z0-9]*[a-z0-9])?(\.[a-z0-9]([-a-z0-9]*[a-z0-9])?)*')

I found a similar issue when the base domain is empty (#59546 (closed)), but then also spec.rules[0].host was invalid. In this case, the base domain is set, everything works fine in review and staging, only for production deployment it does have an empty tls host. I checked the current spec deployed on production and the ingress.yaml shows me:

spec:
  rules:
  - host: example-address.de
    http:
      paths:
      - backend:
          serviceName: production-auto-deploy
          servicePort: 5000
        path: /
  - host: le.example-address.de
    http:
      paths:
      - backend:
          serviceName: production-auto-deploy
          servicePort: 5000
        path: /
  tls:
  - hosts:
    - le.example-address.de
    - api.example-address.de
    secretName: production-auto-deploy-tls

In this log the actual domain setup in CI/CD->Kubernetes->Base domain is swapped to example-address.de

Steps to reproduce

Unclear how it is produced. Cert-Manager and Prometheus is installed and running on GKE Kubernetes with version 1.12.7-gke.24.

Example Project

What is the current bug behavior?

Deploy fails in production.

What is the expected correct behavior?

Successful deployment.

Output of checks

This bug happens on GitLab.com (Silver Plan)