nginx / nginx-ingress-helm-operator

NGINX Ingress Operator for NGINX and NGINX Plus Ingress Controllers. Based on the Helm chart for NGINX Ingress Controller - https://github.com/nginxinc/helm-charts

Repository from Github https://github.comnginx/nginx-ingress-helm-operatorRepository from Github https://github.comnginx/nginx-ingress-helm-operator

Controller's argument does not work with operator v2.3.2 and controller v3.6.2

yosukeora opened this issue · comments

When I released custom resource NginxIngress with current default configuration, with operator version 2.3.2, it ended up in controller's runtime error.

This helm template line https://github.com/nginxinc/nginx-ingress-helm-operator/blob/a9b3457ce42f8d2ac0bfb71f1a17c611a275109d/helm-charts/nginx-ingress/templates/_helpers.tpl#L247

turns to

-v=1

by default.

it causes this error output from the controller pod.

flag provided but not defined: -v
Usage of /nginx-ingress:
NGINX Ingress Controller Version=4.0.0 Commit=b5b1884922d2f1f3f5291edc8c9bb847bb4b31bb Date=2024-11-06T15:16:26Z DirtyState=false Arch=linux/arm64 Go=go1.23.2

This indicates that option -v should be string like 'debug' but operator's definition only allows integer and can't override with controller.logLevel. I already confirmed -v=debug works individually when its set to controller.

Also, weird thing is the error shows controller's version is 4.0.0, while NginxIngress generated Deployment with label app.kubernetes.io/version=3.6.2.

I am trying this on my local minikube environment, so I am checking minikube's ingress addon but its version is v1.10.1.

Anyone encoutered the same situation?

Sorry, my bad!
I set the image version to nginx/nginx-ingress:edge which needed to be nginx/nginx-ingress:3.6.2 on NginxIngress manifest.