Possible issue with partial route matches. Setting an ingress rule for /status will proxy all routes matching /status* to the associated endpoint. This may cause unexpected behavior as shown below.

ingress_partial_route_match

$ kubectl describe svc nginx-plus-service
Name:			nginx-plus-service
Namespace:		default
Labels:			<none>
Selector:		app=nginx-plus-ingress
Type:			ClusterIP
IP:			None
Port:			<unnamed>	8080/TCP
Endpoints:		172.16.81.36:8080
Session Affinity:	None
No events.


$ kubectl get ing
NAME            RULE          BACKEND   ADDRESS
nginx-ingress   -                       
                foo.bar.com   
                /status       nginx-plus-service:8080

$ curl <ingress-host-ip>/status
--> If you read through this, you see 4xx return codes for this service is 0
... "default-nginx-ingress-foo.bar.com-nginx-plus-service" --> "responses":{"1xx":0,"2xx":1,"3xx":0,"4xx":0,"5xx":0,"total":1} ...

$ curl <ingress-host-ip>/status.html
<returns nginx-plus dashboard html, although we may have been looking retrieve a static file for ingress rule / >

$ curl <ingress-host-ip>/status-I-made-this-up
<returns 404 not found **from the service endpoint**, not the default-http-backend being used to serve 404 responses for missed rule matches>

$ curl <ingress-host-ip>/status
--> 4xx return codes reflected on service endpoint.
... "default-nginx-ingress-foo.bar.com-nginx-plus-service" --> "responses":{"1xx":0,"2xx":2,"3xx":0,"4xx":1,"5xx":0,"total":3} ...