The ingress-nginx controller crashes due to unknown directive "alias" when using global snippets.

[albgus]

What happened:
In an attempt to find a workaround for
#9211 #8376 I have added the following snippets to the ingress-nginx configMap.

  http-snippet: |
    error_page 400 /_error/400.html;
    error_page 401 /_error/401.html;
    error_page 403 /_error/403.html;
    error_page 404 /_error/404.html;
    error_page 405 /_error/405.html;
    error_page 408 /_error/408.html;
    error_page 413 /_error/413.html;
    error_page 414 /_error/414.html;
    error_page 416 /_error/416.html;
    error_page 494 /_error/494.html;
    error_page 495 /_error/495.html;
    error_page 496 /_error/496.html;
    error_page 500 /_error/500.html;
    error_page 502 /_error/502.html;
    error_page 503 /_error/503.html;
    error_page 504 /_error/504.html;
  server-snippet: |
    location /_error/ {
        internal;
        client_max_body_size 0;
        alias /etc/nginx/error_pages;
    }

This results in ingress-nginx returning an error stating that the alias directive doesn't exist.

-------------------------------------------------------------------------------
E0429 07:40:57.625086       7 queue.go:131] "requeuing" err=<
	-------------------------------------------------------------------------------
	Error: exit status 1
	2024/04/29 07:40:57 [emerg] 41#41: unknown directive "alias" in /tmp/nginx/nginx-cfg3303928091:461
	nginx: [emerg] unknown directive "root" in /tmp/nginx/nginx-cfg3303928091:461
	nginx: configuration file /tmp/nginx/nginx-cfg3303928091 test failed
	-------------------------------------------------------------------------------

What you expected to happen:
As the alias and root directives are part of the nginx ngx_http_core_module I would have expected them to work.

NGINX Ingress controller version

-------------------------------------------------------------------------------
NGINX Ingress controller
  Release:       v1.10.0
  Build:         71f78d49f0a496c31d4c19f095469f3f23900f8a
  Repository:    https://github.com/kubernetes/ingress-nginx
  nginx version: nginx/1.25.3

-------------------------------------------------------------------------------

Kubernetes version (use kubectl version):

Client Version: v1.29.3
Kustomize Version: v5.0.4-0.20230601165947-6ce0bf390ce3
Server Version: v1.28.1

Environment:

• Cloud provider or hardware configuration:
• OS (e.g. from /etc/os-release):
• Kernel (e.g. uname -a):
• Install tools:
  • Please mention how/where was the cluster created like kubeadm/kops/minikube/kind etc.
• Basic cluster related info:
  • kubectl version
  • kubectl get nodes -o wide

How was the ingress-nginx-controller installed:
Helm chart v4.10.0 (through ArgoCD)

values.yml

controller:
  kind: Deployment
  replicaCount: 2
  config:
    use-forwarded-headers: "true"
    proxy-real-ip-cidr: "10.160.20.0/24"
    log-format-upstream: >-
      $remote_addr - $remote_user [$time_iso8601] "$request" $status $body_bytes_sent
      "$http_referer" "$http_user_agent" $request_length $request_time
      [$namespace/$service_name:$service_port] [$proxy_alternative_upstream_name]
      $upstream_addr $upstream_response_length $upstream_response_time $upstream_status $req_id
    # Tweaks to allow large headers
    large-client-header-buffers: 8 64k
    client-header-buffer-size: 8k
    proxy-buffers-number: 16
    proxy-buffer-size: 32k
    server-snippet: |
      location /_error/ {
          internal;
          client_max_body_size 0;
          alias /etc/nginx/error_pages;
      }
    http-snippet: |
      error_page 400 /_error/400.html;
      error_page 401 /_error/401.html;
      error_page 403 /_error/403.html;
      error_page 404 /_error/404.html;
      error_page 405 /_error/405.html;
      error_page 408 /_error/408.html;
      error_page 413 /_error/413.html;
      error_page 414 /_error/414.html;
      error_page 416 /_error/416.html;
      error_page 494 /_error/494.html;
      error_page 495 /_error/495.html;
      error_page 496 /_error/496.html;
      error_page 500 /_error/500.html;
      error_page 502 /_error/502.html;
      error_page 503 /_error/503.html;
      error_page 504 /_error/504.html;
  allowSnippetAnnotations: true
  extraVolumes:
    - name: nginx-error-pages
      configMap:
        name: ingress-nginx-error-pages
  extraVolumeMounts:
    - name: nginx-error-pages
      mountPath: /etc/nginx/error_pages
  extraArgs:
    default-ssl-certificate: "ingress-nginx/staging-internal-default-tls"
  watchIngressWithoutClass: false
  ingressClassResource:
    name: internal-nginx
    enabled: true
    default: true
    controllerValue: "k8s.io/internal-nginx"
  ingressClass: internal-nginx
  podAnnotations:
    prometheus.io/scrape: true
    prometheus.io/port: 10254
  metrics:
    enabled: true
  service:
    type: NodePort
    nodePorts:
      http: 30080
      https: 30443
  resources:
    requests:
      cpu: 50m
      memory: 256Mi
    limits:
      memory: 256Mi
  updateStrategy:
    type: RollingUpdate
    rollingUpdate:
      maxSurge: 1
      maxUnavailable: 1

**How to reproduce this issue**:

Anything else we need to know:

[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/test-infra repository.

[longwuyuan]

/remove-kind bug
/kind support

Try the root directive https://nginx.org/en/docs/http/ngx_http_core_module.html#alias

The nginx directives you choose are not coded in the controller, just to state the obvious.

Please feel free to re-open the issue and add the details as asked in a new bug report template, in case you find problem in the controller itself.

/close

[k8s-ci-robot]

@longwuyuan: Closing this issue.

In response to this:

/remove-kind bug
/kind support

Try the root directive https://nginx.org/en/docs/http/ngx_http_core_module.html#alias

The nginx directives you choose are not coded in the controller, just to state the obvious.

Please feel free to re-open the issue and add the details as asked in a new bug report template, in case you find problem in the controller itself.

/close

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/test-infra repository.

[longwuyuan]

Actually I was wrong earlier (or maybe not as you are using snippet.sorry not clear) It seems the project removed the support for both alias and root directive #8624

Example of a custom backend service with custom error pages https://kubernetes.github.io/ingress-nginx/examples/customization/custom-errors/

[albgus]

Ok, that is a bit unfortunate. As mentioned in my original issue the documented method doesn't work because it enforces overwriting all errors returned from the backend, which is not what I want. However perhaps it's possible to implement something similar to the included error handling with a custom backend.