K
Q

How to configure nginx when using Helm to install?

1/11/2019

According to this article one need to set some nginx properties when running a .NET Core 2.x application using https and Azure AD Authentication behind Nginx in a Kubernetes cluster:

.Net Core behind NGINX returns 502 Bad Gateway after authentication by IdentityServer4

The answer outlines how to do this for a regular Nginx installation, but I would like to do this when installing Nginx in a Kubernetes cluster using Helm.

These are the Nginx properties one need to set:

nginx.conf:
    http{
    ...
    proxy_buffer_size   128k;
    proxy_buffers   4 256k;
    proxy_busy_buffers_size   256k;
    large_client_header_buffers 4 16k;
    ...
    }


default.conf:
location /{
    ...
    fastcgi_buffers 16 16k;
    fastcgi_buffer_size 32k;
    ...
}

The command I use to install Nginx in the Kubernetes cluster is:

helm install stable/nginx-ingress --namespace kube-system

How does one set the above properties when installing Nginx using Helm in a Kubernetes cluster?

-- OlavT
.net
asp.net-core
kubernetes
kubernetes-helm
nginx

Similar Questions

Maven and Docker Builds, Releases and Deploy on Kubernetes/OpenShift
Adding an exception to kubernetes garbage collection
Go command to get updated output using kubectl
TLS version and ciphers in Istio mTLS

3 Answers

1/3/2020

I found that the simplest solution is to use annotations on the Nginx ingress like this:

apiVersion: extensions/v1beta1
kind: Ingress
metadata:
  name: test-ingress
  annotations:
    kubernetes.io/ingress.class: nginx
    nginx.ingress.kubernetes.io/proxy-buffer-size: "16k"
    nginx.ingress.kubernetes.io/rewrite-target: /
...
-- OlavT
Source: StackOverflow
1/15/2019

It is possible to customize nginx configuration using ConfigMap.
This is an example
after adding the ConfigMap if nginx doesn't get updated, update your nginx chart using Helm:
helm upgrade [RELEASE] [CHART]
heml upgrade my-release stable/nginx-ingress
if you don't know the [RELEASE] use following command:
helm list

-- Mozafar Gholami
Source: StackOverflow
1/15/2019

Fully agree with @Mozafar Gholami, you can change parameters using ConfigMap while deploying nginx or update your current configuration. To update parameters before installation suggest you next:

1.fetch chart to your local machine and unzip it:

helm fetch stable/nginx-ingress

tar -xzf nginx-ingress-1.1.4.tgz
  1. edit controller.config section in the values.yml

example for you:

controller:
  name: controller
  image:
    repository: quay.io/kubernetes-ingress-controller/nginx-ingress-controller
    tag: "0.21.0"
    pullPolicy: IfNotPresent
    # www-data -> uid 33
    runAsUser: 33

  config:
    proxy-buffer-size: "128k"
    proxy-buffers: "4 256k"

3.check what will be added to new configmap

helm   template . | less

4. install chart

helm install --name nginx-ingress --namespace kube-system ./nginx-ingress

Please keep in mind that:

  1. Instead of ConfigMaps you can change parameters with Annotations.

  2. Unfortunately NOT ALL parameters can be changed in nginx-ingress by above approach.

  3. For more information reading the nginx-ingress customization page where you can find all values you are able to change. For example in your case I wasnt able to update proxy_busy_buffers_size and large_client_header_buffers parameters.

Hope this help you.

-- VKR
Source: StackOverflow