Add an auxiliary configuration

For the most part, the ingress controller will handle the underlying HAProxy configuration file and you only need to set annotations on your Ingress, Service and ConfigMap resources. However, it’s still possible to write raw HAProxy configuration directives for features not yet covered by annotations.

One option is to use the following annotations:

• backend-config-snippet
• frontend-config-snippet
• global-config-snippet
• stats-config-snippet

You can also provide an auxiliary configuration file that defines entirely new config sections, such as to define cache, mailers, or ring sections. This file will be loaded when the ingress controller starts up.

You can thus:

• Configure anything not supported by Ingress Controller annotations.
• Migrate a legacy HAProxy Enterprise configuration to the HAProxy Enterprise Kubernetes Ingress Controller.

Primary and auxiliary configuration files

Here are the differences between the two types of configuration files:

Deploy the Ingress controller

1. Create the namespace haproxy-controller if it does not exist:

   $ kubectl create namespace haproxy-controller

2. Add the HAProxy configuration directives that you want to use to a new .cfg file. Below, we define a cache section in a file named haproxy-auxiliary.cfg:

   cache mycache
     total-max-size 4095
     max-object-size 10000
     max-age 30
   

   Be sure to add a blank line at the end of the file, otherwise HAProxy will get the error Missing LF on last line, file might have been truncated and fail to load the file.

3. Load the file into your Kubernetes cluster as a ConfigMap resource.

   $ kubectl create configmap haproxy-auxiliary-configmap \
       --from-file haproxy-auxiliary.cfg \
       --namespace haproxy-controller

   • A ConfigMap is an API object for storing non-confidential information in key-value pairs that makes your applications easily portable.
   • Kubernetes automatically updates the mounted volume when you update the ConfigMap.

4. Deploy the Ingress controller again with the ConfigMap attached as a ConfigMap volume. Below we show an example using Helm.

   a. Create a file named values.yaml that contains the following:

   controller:
     extraVolumes:
     - name: haproxy-auxiliary-volume
       configMap:
         name: haproxy-auxiliary-configmap
     extraVolumeMounts:
     - name: haproxy-auxiliary-volume
       mountPath: /usr/local/etc/haproxy/haproxy-aux.cfg
       subPath: haproxy-auxiliary.cfg

   • Note: The mountPath field must be set to /usr/local/etc/haproxy/haproxy-aux.cfg for both the Community and Enterprise versions.

   b. Uninstall the Helm chart if you installed it previously.

   $ helm uninstall haproxy-kubernetes-ingress --namespace haproxy-controller

   c. Install the Helm chart, referencing the values.yaml file with the -f flag.

   $ helm install haproxy-kubernetes-ingress haproxytech/kubernetes-ingress \
       -f values.yaml \
       --namespace haproxy-controller

5. To make use of the cache section, you could add a backend-config-snippet annotation to a Service resource:

   apiVersion: v1
   kind: Service
   metadata:
     labels:
       run: api
     name: api
     annotations:
       haproxy.org/backend-config-snippet: |
         http-request cache-use mycache
         http-response cache-store mycache

See also

• ConfigMaps
• Extra volumes

Update the configuration file

Follow these steps to update the auxiliary configuration file later.

1. Call kubectl edit to edit the ConfigMap.

   $ kubectl edit configmap haproxy-auxiliary-configmap --namespace haproxy-controller

2. The file opens in your text editor. Make changes to the file and then save and close it. The Ingress controller will detect the change and reload the file.

Next up