Nginx Throttle

A simple Docker container with Nginx reverse proxy for Apps that are behind an AWS ELB and need to throttle HTTP requests.

Reference: http://nginx.org/en/docs/http/ngx_http_limit_req_module.html

Usage

Inside a POD file

apiVersion: v1
kind: Pod
metadata:
  name: myapp
spec:
  hostNetwork: true
  containers:
  - name: myapp
    image: myapp-image
    env:
      - name: MY_APP_VAR
        value: "my app value"
    ports:
    - containerPort: 3000
      protocol: TCP
    livenessProbe:
      failureThreshold: 3
      httpGet:
        path: /health
        port: 3000
        scheme: HTTP
      initialDelaySeconds: 30
      periodSeconds: 10
      successThreshold: 1
      timeoutSeconds: 5
  - name: nginx
    image: vivareal/nginx-throttle:latest
    env:
      - name: NGINX_BACKEND_ENDPOINT
        value: "localhost:3000"
      - name: NGINX_THROTTLE_RATE_LIMIT
        value: "200r/s"
      - name: NGINX_SET_NODELAY
        value: "true"
    ports:
    - containerPort: 80
      protocol: TCP
    livenessProbe:
      failureThreshold: 3
      httpGet:
        path: /health
        port: 80
        scheme: HTTP
      initialDelaySeconds: 30
      periodSeconds: 10
      successThreshold: 1
      timeoutSeconds: 5

Docker compose

version: '2'

services:
   myapp:
     image: myapp-image
     restart: always
     ports:
       - "3000:3000"
     environment:
       MY_APP_VAR: "my app value"

   nginx:
     depends_on:
       - myapp
     image: vivareal/nginx-throttle:latest
     ports:
       - "80:80"
     environment:
       NGINX_BACKEND_ENDPOINT: "myapp:3000"
       NGINX_THROTTLE_RATE_LIMIT: "200r/s"
       NGINX_SET_NODELAY: true

Just Docker

Start your app:

docker run -d --name=myapp -p 3000:3000 myapp-image

Now start the nginx-throttle with a link to your app:

docker run -d --name=nginx --link myapp -p 80:80  -e NGINX_THROTTLE_RATE_LIMIT="200r/m" -e NGINX_BACKEND_ENDPOINT="myapp:3000" -e NGINX_SET_NODELAY=true vivareal/nginx-throttle:latest

Configuration

All settings are passed by environment variables:

Variable	Default value	Required
NGINX_BACKEND_ENDPOINT	-	YES
NGINX_LISTEN_PORT	80	No
NGINX_PROXY_TIMEOUT	90	No
NGINX_SET_NODELAY	false	No
NGINX_THROTTLE_BURST	10	No
NGINX_THROTTLE_MEM_LIMIT	10m	No
NGINX_THROTTLE_RATE_LIMIT	100r/s	No
NGINX_WORKER_CONNECTIONS	4096	No

Details

NGINX_BACKEND_ENDPOINT

• This is a required setting.
• Value type is string.
• There is no default value for this setting.

Set the address of your app for Nginx proxy pass. The format is <addr>:<port> E.g.: localhost:3000

NGINX_LISTEN_PORT

• Value type is number.
• Default value is 80.

Set the port where Nginx should listen on.

NGINX_PROXY_TIMEOUT

• Value type is number.
• Default value is 90.

Number in seconds for proxy timeout.

NGINX_SET_NODELAY

• Value type is boolean.
• Default value is false.

Set to true if you want Nginx to drop requests instead of delaying excessive requests.

NGINX_THROTTLE_BURST

• Value type is number.
• Default value is 10.

Excessive requests are delayed until their number exceeds the maximum burst size in which case the request is terminated with an error 503 (Service Temporarily Unavailable). This Environment variable sets the size of the burst.

NGINX_THROTTLE_MEM_LIMIT

• Value type is string.
• Default value is 10m.

Set the size of memory where the states are kept. In particular, the state stores the current number of excessive requests.

NGINX_THROTTLE_RATE_LIMIT

• Value type is string.
• Default value is 100r/s.

Set the maximum rate of requests. The rate is specified in requests per second (r/s). If a rate of less than one request per second is desired, it is specified in request per minute (r/m). For example, half-request per second is 30r/m.

NGINX_WORKER_CONNECTIONS

• Value type is number.
• Default value is 4096.

Set the number of connections that nginx worker can handle.