This side effect is introduced by
Basically it changes the jenkinsio containers to redirect every non jenkins.io request to https://jenkins.io/<uri>.
What happened here, is that kubelet send http healthcheck to container without using jenkins.io url, so the query was redirected externally through Ingress -> Service -> Container but because no containers were attached to the jenkinsio service (healt check wasn't successful yet), the redirection was failing.
In order to fix this we have two solutions.
1) Modify jenkins.io containers to only redirect jenkins-ci.org,www.jenkins-ci.org,www.jenkins.io to jenkins.io
2) Health check monitor port instead of HTTP url
I am more in favor of solution two as the health check doesn't rely on website content but only if Nginx started or not.