Description
What happened:
When defining two backends, as such:
- backend:
service:
name: service-a
port:
number: 80
path: /
pathType: Exact
- backend:
service:
name: service-b
port:
number: 80
path: /
pathType: Prefix
We only get metrics for the service-a backend.
What you expected to happen:
To be able to get metrics for each backend.
NGINX Ingress controller version (exec into the pod and run nginx-ingress-controller --version.):
NGINX Ingress controller
Release: v1.3.0
Build: 2b7b74854d90ad9b4b96a5011b9e8b67d20bfb8f
Repository: https://github.com/kubernetes/ingress-nginx
nginx version: nginx/1.19.10
Kubernetes version (use kubectl version):
Server Version: version.Info{Major:"1", Minor:"25", GitVersion:"v1.25.8", GitCommit:"0ce7342c984110dfc93657d64df5dc3b2c0d1fe9", GitTreeState:"clean", BuildDate:"2023-03-15T13:33:02Z", GoVersion:"go1.19.7", Compiler:"gc", Platform:"linux/amd64"}
Environment:
-
Cloud provider or hardware configuration: AWS (via kOps)
-
OS (e.g. from /etc/os-release): Ubuntu 20.04.5 LTS
-
Kernel (e.g.
uname -a): 5.15.0-1031-aws -
Install tools: kOps
Please mention how/where was the cluster created like kubeadm/kops/minikube/kind etc.
-
Basic cluster related info:
-
kubectl versionServer Version: version.Info{Major:"1", Minor:"25", GitVersion:"v1.25.8", GitCommit:"0ce7342c984110dfc93657d64df5dc3b2c0d1fe9", GitTreeState:"clean", BuildDate:"2023-03-15T13:33:02Z", GoVersion:"go1.19.7", Compiler:"gc", Platform:"linux/amd64"} -
kubectl get nodes -o wideUbuntu 20.04.5 LTS 5.15.0-1031-aws containerd://1.6.18
-
-
How was the ingress-nginx-controller installed:
Installed via Helm using ArgoCD -
Current State of the controller:
Controller is running and working as expected, except for this metric -
Current state of ingress object, if applicable:
All Ingress objects are fine and the controller handles requests fine.
How to reproduce this issue:
Create an ingress object with multiple / paths, of different types.
Send traffic to both matches / and /test.
Look at resulting metrics.
Anything else we need to know:
I looked at the resulting configuration and compared them to see if I can find the issue. This seems to highlight the bug:
$ diff -u a b
--- a 2023-07-13 10:56:34
+++ b 2023-07-13 10:56:42
@@ -1,4 +1,4 @@
- location / {
+ location = / {
set $namespace "mynamespace";
set $ingress_name "myingress";
@@ -46,7 +46,7 @@
port_in_redirect off;
set $balancer_ewma_score -1;
- set $proxy_upstream_name "mynamespace-service-a-80";
+ set $proxy_upstream_name "mynamespace-service-b-80";
set $proxy_host $proxy_upstream_name;
set $pass_access_scheme $scheme;
This config is fine, except that set $service_name is the same in both.
It seems that the problem is happening around here:
- https://github.com/kubernetes/ingress-nginx/blob/controller-v1.3.0/rootfs/etc/nginx/template/nginx.tmpl#L1246
- https://github.com/kubernetes/ingress-nginx/blob/controller-v1.3.0/rootfs/etc/nginx/template/nginx.tmpl#L1189
Which I think comes from https://github.com/kubernetes/ingress-nginx/blob/controller-v1.3.0/internal/ingress/controller/template/template.go#L1061-L1149
I think the expected diff is as such:
$ diff -u a b
--- a 2023-07-18 13:13:39
+++ b 2023-07-13 10:56:42
@@ -1,8 +1,8 @@
- location / {
+ location = / {
set $namespace "mynamespace";
set $ingress_name "myingress";
- set $service_name "service-a";
+ set $service_name "service-b";
set $service_port "80";
set $location_path "/";
set $global_rate_limit_exceeding n;
@@ -46,7 +46,7 @@
port_in_redirect off;
set $balancer_ewma_score -1;
- set $proxy_upstream_name "mynamespace-service-a-80";
+ set $proxy_upstream_name "mynamespace-service-b-80";
set $proxy_host $proxy_upstream_name;
set $pass_access_scheme $scheme;
Metadata
Metadata
Assignees
Labels
Type
Projects
Status
No status