Experiencing 500 errors when trying to use defaultBackend with scope (on the controller)

[DanielRLittle]

What happened:

When adding a default backend to the nginx controller when the controller has a namespace defined in scope, I am receiving a 500 internal server error. When removing the namespace from the scope, the default backend and custom http errors work fine. There is network connectivity between the controller and the default backend, as I can curl the service endpoint from the pods and due to the fact that it works when scope is disabled.

What works (shows custom http errors): controller: config:

• custom-http-errors: "404" scope:
• enabled: false defaultBackend:
  • enabled: true

What doesn't work (raises a 500): controller: config:

• custom-http-errors: "404" scope:
• enabled: false defaultBackend:
  • enabled: true

What you expected to happen:

I would expect the custom http errors to still work when scope is enabled for a namespace

NGINX Ingress controller version:

v1.5.1

Kubernetes version:

Client version: v1.22.0 Server Version: v1.25.11

Environment:

• Cloud provider or hardware configuration: EKS

• OS (e.g. from /etc/os-release): NAME="Alpine Linux" ID=alpine VERSION_ID=3.17.2 PRETTY_NAME="Alpine Linux v3.17"

• Kernel (e.g. uname -a): Linux ingress-nginx-controller-54788bcbfc-4ks8x 5.10.184-175.749.amzn2.x86_64 #1 SMP Wed Jul 12 18:40:28 UTC 2023 x86_64 Linux

• Install tools:

  • gitlab ci, and locally via debian container

• How was the ingress-nginx-controller installed:

  • If helm was used then please show output of helm ls -A | grep -i ingress ingress-nginx ingress-nginx 220 2023-08-11 10:04:27.644750249 +0000 UTC deployed ingress-nginx-4.4.0 1.5.1

How to reproduce this issue:

• Add a default backend to an nginx controller
• Get it working without scope
• Add scope with a dfifferent namespace
• Hopefully start experiencing 500 errors

[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](https://git.k8s.io/community/contributors/guide/pull-requests.md). If you have questions or suggestions related to my behavior, please file an issue against the [kubernetes/test-infra](https://github.com/kubernetes/test-infra/issues/new?title=Prow%20issue:) repository.

[longwuyuan]

Namespace scoping is work in progress but the controller has evolved in way that other users configure specifics to get isolation per controller.

THat being said, the description here can get better to help someone reproduce he problem. Please write a step-by-step guide so someone can copy/paste from it and reproduce the problem.

/remove-kind bug

[github-actions[bot]]

This is stale, but we won't close it automatically, just bare in mind the maintainers may be busy with other tasks and will reach your issue ASAP. If you have any question or request to prioritize this, please reach #ingress-nginx-dev on Kubernetes Slack.