Configure HTTP Load Balancing

Learn how to configure HTTP load balancing traffic to several instances of an app.

This topic describes an early access feature. These features are provided for you to try before they are generally available. You shouldn't use early access features for production purposes.

Share your thoughts about the documentation by taking this short survey: https://forms.gle/w6eDi53F4UcmDjw68.

Objective

The App team wants to provide external access to their custom app which has two instances running on their internal network. External access will be provided from http://www.example.com. The two instances of their app are available at the following internal IP addresses:

http://198.51.100.0:8080
http://198.51.100.0:8081

Prerequisites

The environment has been set up as follows for this use case:

The Platform team has set up NGINX Management Suite permissions to allow the App team to create any App Delivery Manager related resource.
The Platform team has created an instance group, example-ig, for the NGINX instances that will be used to route the traffic to the team’s app instances (workloads, in App Delivery Manager terminology, which typically map to upstreams in NGINX terminology).

Solution

How to Access the User Interface

This guide provides instructions for completing tasks using the NGINX Management Suite App Delivery Manager web interface.

To access the web interface, go to the FQDN of your NGINX Management Suite host and log in. On the Launchpad menu, select “App Delivery Manager.”

Create an Environment

If you haven’t already, create an Environment resource by following these steps:

Select Environments on the App Delivery Manager section of the sidebar. The list of existing environments will be displayed.
Select Create Environment. In the Create Environment drawer, type Tutorial Environment in the Name field. You can leave the default values for all the other fields (Note: This tutorial does not require customized templates).
Select Submit to create the environment.

Create a Gateway

The gateway will describe how traffic will be routed through NGINX instances to get to the actual app workloads.

Select Gateways on the sidebar. The list of existing gateways will be displayed.
Select Create Gateway. In the Configuration section of the Create Gateway drawer type Example Gateway in the Name field.
Select Tutorial Environment in the Environment list.
Select Next to go to the Placements section.
In the Instance Group Refs list, select example-ig (Which was created by the Platform team as part of the prerequisites), then select Done.
Select Next to continue to the Hostnames page.
Type http://www.example.com in the Hostname field, and then Select Done.
Select Submit to create the gateway (Note: To create a secure website, you would need to type an HTTPS address and specify the cert to use).

Create an App

Follow these steps to create the app:

Select Apps on the sidebar. The list of existing apps will be displayed.
Select Create App.
In the Create App drawer, type Example App in the Name field.
Select Tutorial Environment in the Environment list. You can leave the default values for all the other fields.
Select Submit to create the app.

Create a Web Component

The app we just created is a wrapper that can be composed of multiple components, each potentially referencing a unique microservice. For our tutorial, we are deploying a simple app with only one component. To create this component, take the following steps:

If you are not on Apps Overview page already, select Apps on the sidebar. The list of existing apps will be displayed.
On the Overview page, in the Name column, select the Example App link in the list of apps.
At the top of the page, select Web Components.
On the Web Components page, select Create Web Component.
On the Configuration section of the Create Web Component drawer, type Example Component in the Name field.
In the Gateway Refs field list, select Example Gateway.
Select Next to advance to the URIs page.
Type / in the URI field (If you are not able to type a value, select the pencil icon to edit the URI).
Select Next to continue to the Workload Groups page.
Select the Add Workload Group button.
Type wg1 in the Workload Group Name field.
In the Backend Workload URIs section, type http://198.51.100.0:8080 in the URI field. Select Done.
Select Add Backend Workload URI to add another workload. Type http://198.51.100.0:8081 in the URI field. Then select Done.
In the Workload Groups page, select Done.
Select Submit to complete the component configuration.

Resulting NGINX Configuration

After completing the steps above, the web component will transition into a “Configured” state. If you examine any of the instances that belong to the example-ig instance group, you will see a configuration similar to the following:

server {
    server_name www.example.com;
    listen 80 reuseport;
    status_zone b57757a6-ef8b-3ef0-be2a-067d66360680;
    f5_metrics_marker environment ca6aa2ef-717f-48ab-96d9-fc1b79e3ec43;
    f5_metrics_marker gateway 4b727d89-b15d-45c1-8bf3-8b6a62fda9ac;
    # Generated by web component Example Component(e0410d7c-d6d4-442d-840f-6d7ad30b5445)
    location / {
        proxy_set_header X-Forwarded-For $remote_addr;
        proxy_set_header Host $host
        proxy_set_header Connection "";
        proxy_http_version 1.1;
        proxy_pass http://wg1_http_e0410d7c-d6d4-442d-840f-6d7ad30b5445;
        # metrics
        f5_metrics_marker app e2f3d17d-929a-4ee0-90b2-55fd96f66dfe;
        f5_metrics_marker component e0410d7c-d6d4-442d-840f-6d7ad30b5445;
    }
}
upstream wg1_http_e0410d7c-d6d4-442d-840f-6d7ad30b5445 {
    zone wg1_e0410d7c-d6d4-442d-840f-6d7ad30b5445 1280K;
    server 198.51.100.0:8080;
    server 198.51.100.0:8081;
}