Hello. This is probably down to failing to read the right end of the manual.
My naive assumption is that I can helm-install kong with ingress-controller enabled, and as a basic test modify a working ingress config in another helm chart, change the ingress annotation to ‘kong’, have the ingress controllers fight it out amongst themselves and then observe that kong handles that service now.
Indeed, the service & route turn up in the kong config. However, the service hostname now has ‘.default.svc’ appended. This may (or may not) have explain why hitting that service’s ingress hostname results in a 404.
What dreadfully obvious thing have I failed to spot?
I’m sure there’s a better way than ‘configure with empty loadBalancerIP, promote the external IP with the forwarding rule to ‘static’ and edit the helm chart accordingly.’ But that’s what seems to work, so…