このページは、まだ日本語ではご利用いただけません。翻訳中です。

Rate-Limiting

Kong can rate limit traffic without any external dependency. Kong stores the request counters in-memory and each Kong node applies the rate limiting policy independently without synchronization of information. However, if Redis is available in your cluster, Kong can take advantage of it and synchronize the rate limit information across multiple Kong nodes and enforce a slightly different rate limiting policy. You can use Redis for rate limiting in a multi-node Kong deployment.

You can use the Kong Gateway Enterprise Secrets Management feature along with the example rate-limiting plugin. If you have an existing plugin that you wish to use Secrets Management with, you can skip directly to the Secrets Management section and use it for your plugin instead of the rate-limiting plugin.

Deploy an echo service

To proxy requests, you need an upstream application to send a request to. Deploying this echo server provides a simple application that returns information about the Pod it’s running in:

kubectl apply -f https://docs.jp.konghq.com/assets/kubernetes-ingress-controller/examples/echo-service.yaml

The results should look like this:

service/echo created
deployment.apps/echo created

Add routing configuration

Create routing configuration to proxy /echo requests to the echo server:

echo "
apiVersion: gateway.networking.k8s.io/v1
kind: HTTPRoute
metadata:
  name: echo
  annotations:
    konghq.com/strip-path: 'true'
spec:
  parentRefs:
  - name: kong
  rules:
  - matches:
    - path:
        type: PathPrefix
        value: /echo
    backendRefs:
    - name: echo
      kind: Service
      port: 1027
" | kubectl apply -f -

echo "
apiVersion: networking.k8s.io/v1
kind: Ingress
metadata:
  name: echo
  annotations:
    konghq.com/strip-path: 'true'
spec:
  ingressClassName: kong
  rules:
  - http:
      paths:
      - path: /echo
        pathType: ImplementationSpecific
        backend:
          service:
            name: echo
            port:
              number: 1027
" | kubectl apply -f -

The results should look like this:

httproute.gateway.networking.k8s.io/echo created

ingress.networking.k8s.io/echo created

Test the routing rule:

curl -i $PROXY_IP/echo

The results should look like this:

HTTP/1.1 200 OK
Content-Type: text/plain; charset=utf-8
Content-Length: 140
Connection: keep-alive
Date: Fri, 21 Apr 2023 12:24:55 GMT
X-Kong-Upstream-Latency: 0
X-Kong-Proxy-Latency: 1
Via: kong/3.2.2

Welcome, you are connected to node docker-desktop.
Running on Pod echo-7f87468b8c-tzzv6.
In namespace default.
With IP address 10.1.0.237.
...

If everything is deployed correctly, you should see the above response. This verifies that Kong Gateway can correctly route traffic to an application running inside Kubernetes.

Set up rate limiting

1. Create an instance of the rate-limiting plugin.

    echo "
    apiVersion: configuration.konghq.com/v1
    kind: KongPlugin
    metadata:
      name: rate-limit
      annotations:
        kubernetes.io/ingress.class: kong
    config:
      minute: 5
      policy: local
    plugin: rate-limiting
    " | kubectl apply -f -
   

   The results should look like this:

    kongplugin.configuration.konghq.com/rate-limit created
   

2. Associate the plugin with the Service.

    kubectl annotate service echo konghq.com/plugins=rate-limit
   

   The results should look like this:

    service/echo annotated
   

3. Send requests through this Service to rate limiting response headers.

    curl -si $PROXY_IP/echo | grep RateLimit
   

   The results should look like this:

    RateLimit-Limit: 5
    X-RateLimit-Remaining-Minute: 4
    X-RateLimit-Limit-Minute: 5
    RateLimit-Reset: 60
    RateLimit-Remaining: 4
   

4. Send repeated requests to decrement the remaining limit headers, and block requests after the fifth request.

    for i in `seq 6`; do curl -sv $PROXY_IP/echo 2>&1 | grep "< HTTP"; done
   

   The results should look like this:

    < HTTP/1.1 200 OK
    < HTTP/1.1 200 OK
    < HTTP/1.1 200 OK
    < HTTP/1.1 200 OK
    < HTTP/1.1 200 OK
    < HTTP/1.1 429 Too Many Requests
   

Scale to multiple pods

1. Scale your Deployment to three replicas, to test with multiple proxy instances.

    kubectl scale --replicas 3 -n kong deployment kong-gateway
   

   The results should look like this:

    deployment.apps/kong-gateway scaled
   

2. Check if the status of all the Pods that are READY is Running using the command kubectl get pods -n kong.

3. Sending requests to this Service does not reliably decrement the remaining counter.

    for i in `seq 10`; do curl -sv $PROXY_IP/echo 2>&1 | grep "X-RateLimit-Remaining-Minute"; done
   

   The results should look like this:

    < X-RateLimit-Remaining-Minute: 4
    < X-RateLimit-Remaining-Minute: 4
    < X-RateLimit-Remaining-Minute: 3
    < X-RateLimit-Remaining-Minute: 4
    < X-RateLimit-Remaining-Minute: 3
    < X-RateLimit-Remaining-Minute: 2
    < X-RateLimit-Remaining-Minute: 3
    < X-RateLimit-Remaining-Minute: 2
    < X-RateLimit-Remaining-Minute: 1
    < X-RateLimit-Remaining-Minute: 1
   

   The policy: local setting in the plugin configuration tracks request counters in each Pod’s local memory separately. Counters are not synchronized across Pods, so clients can send requests past the limit without being throttled if they route through different Pods.

   Using a load balancer that distributes client requests to the same Pod can alleviate this somewhat, but changes to the number of replicas can still disrupt accurate accounting. To consistently enforce the limit, the plugin needs to use a shared set of counters across all Pods. The redis policy can do this when a Redis instance is available.

Deploy Redis to your Kubernetes cluster

Redis provides an external database for Kong components to store shared data, such as rate limiting counters. There are several options to install it:

Bitnami provides a Helm chart for Redis with turnkey options for authentication.

1. Create a password Secret and replace PASSWORD with a password of your choice.

   kubectl create -n kong secret generic redis-password-secret --from-literal=redis-password=PASSWORD
   

   The results should look like this:

   secret/redis-password-secret created
   

2. Install Redis

    helm install -n kong redis oci://registry-1.docker.io/bitnamicharts/redis \
      --set auth.existingSecret=redis-password-secret \
      --set architecture=standalone
   

   Helm displays the instructions that describes the new installation.

3. Update your plugin configuration with the redis policy, Service, and credentials. Replace PASSWORD with the password that you set for Redis.

    kubectl patch kongplugin rate-limit --type json --patch '[
      {
        "op":"replace",
        "path":"/config/policy",
        "value":"redis"
      },
      {
        "op":"add",
        "path":"/config/redis_host",
        "value":"redis-master"
      },
      {
        "op":"add",
        "path":"/config/redis_password",
        "value":"PASSWORD"
      }
    ]'
   

   The results should look like this:

    kongplugin.configuration.konghq.com/rate-limit patched
   

   If the redis_username is not set , it uses the default redis user.

Test rate limiting is a multi-node Kong deployment

Send requests to the Service with rate limiting response headers.

for i in `seq 10`; do curl -sv $PROXY_IP/echo 2>&1 | grep "X-RateLimit-Remaining-Minute"; done

The results should look like this:

< X-RateLimit-Remaining-Minute: 4
< X-RateLimit-Remaining-Minute: 3
< X-RateLimit-Remaining-Minute: 2
< X-RateLimit-Remaining-Minute: 1
< X-RateLimit-Remaining-Minute: 0
< X-RateLimit-Remaining-Minute: 0
< X-RateLimit-Remaining-Minute: 0
< X-RateLimit-Remaining-Minute: 0
< X-RateLimit-Remaining-Minute: 0
< X-RateLimit-Remaining-Minute: 0

The counters decrement sequentially regardless of the Kong Gateway replica count.

(Optional) Use Secrets Management

Secrets Management is a Kong Gateway Enterprise feature for storing sensitive plugin configuration separately from the visible plugin configuration. The rate-limiting plugin supports Secrets Management for its redis_username and redis_password fields.

Secrets Management supports several backend systems. This guide uses the environment variable backend, which requires minimal configuration and integrates well with Kubernetes’ standard Secret-sourced environment variables.

Add environment variable from Secret

Update your proxy Deployment with an environment variable sourced from the redis-password-secret Secret.

kubectl patch deploy -n kong kong-gateway --patch '
{
  "spec": {
    "template": {
      "spec": {
        "containers": [
          {
            "name": "proxy",
            "env": [
              {
                "name": "SECRET_REDIS_PASSWORD",
                "valueFrom": {
                  "secretKeyRef": {
                    "name": "redis-password-secret",
                    "key": "redis-password"
                  }
                }
              }
            ]
          }
        ]
      }
    }
  }
}'

The results should look like this:

deployment.apps/kong-gateway patched

Update the plugin to use a reference

After the vault has an entry, you can use a special {vault://VAULT-TYPE/VAULT-KEY} value in plugin configuration instead of a literal value. Patch the rate-limit KongPlugin to change the redis_password value to a vault reference.

kubectl patch kongplugin rate-limit --type json --patch '[
  {
    "op":"replace",
    "path":"/config/redis_password",
    "value":"{vault://env/secret-redis-password}"
  }
]'

Check plugin configuration

The updated KongPlugin propagates the reference to the proxy configuration. You can confirm it by checking the admin API.

In one terminal, open a port-forward to the admin API:

kubectl port-forward deploy/kong-gateway -n kong 8444:8444

The results should look like this:

Forwarding from 127.0.0.1:8444 -> 8444

In a separate terminal, query the /plugins endpoint and filter out the rate-limiting plugin:

curl -ks https://localhost:8444/plugins/ | jq '.data[] | select(.name="rate-limiting") | .config.redis_password'

The results should look like this:

"{vault://env/secret-redis-password}"