nginx_ingress_controller_requests are having host data as "_"

[ajsmksja]

What happened:

We are using customized ingress-nginx v1.10.1. We observed that for few of **nginx_ingress_controller_requests metrics when parsed are having host= __ What could be the possibility/scenario for ingress-nginx to get the host data as "" instead of actual host details.

What you expected to happen:
nginx_ingress_controller_requests{canary="",controller_class="k8s.io/ingress-nginx",controller_namespace="xxxxxxx",controller_pod="xxxxxxxxxxxxxx",host="xxxxxxxxxxxxxxxxxxxxxxxx",ingress="xxxxx",method="POST",namespace="xxxxxx",path="/",service="xxxxxx",status="200"}

But for few of nginx_ingress_controller_requests we are getting host as below as "_"

nginx_ingress_controller_requests{canary="",controller_class="k8s.io/ingress-nginx",controller_namespace="xxxxxxxx",controller_pod="xxxxxxxxxxxxxxxxxxxxxxx",host="_",ingress="xxxxxxx",method="POST",namespace="xxxxxxxx",path="/",service="xxxxx",status="200"}

NGINX Ingress controller version
NGINX Ingress controller
Release: 1.10.1
Build: git-be46124cc
Repository: https://github.com/kubernetes/ingress-nginx.git
nginx version: nginx/1.21.6

Kubernetes version
Client Version: v1.31.2
Kustomize Version: v5.4.2
Server Version: v1.29.1

[k8s-ci-robot]

This issue is currently awaiting triage.

If Ingress contributors determines this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[longwuyuan]

Its not possible to reproduce problems or test a feature, using such an old and unsupported version of the controller.

[ajsmksja]

Hi @longwuyuan , this was a finding at a server using older version.
latest finding, we are actually facing the same issue with v1.10.1 too and we are in middle of uplifting to v1.11.2
so we believe we are missing something in here.

[sravanith]

Hi Team ,
As per our finding , we are facing this issue only in few servers, so we think this is not version specific issue, We are also using 1.10.1 and v1.11.2 ingress-nginx , we think this issue might get reproduce here as well. We are looking for possible scenario for having host label data as '_" in nginx_ingress_controller_requests metrics