Service to service rate limiting

TSB is capable of applying rate limits for both gateways and sidecars. In this document, we will enable rate limiting for sidecars to control quota for service to service traffic.

Before you get started, make sure you:
✓ Familiarize yourself with TSB concepts
✓ Install the TSB environment. You can use TSB demo for quick install
✓ Completed TSB usage quickstart. This document assumes you already created Tenant and are familiar with Workspace and Config Groups. Also you need to configure tctl to your TSB environment.

Enable Rate Limiting Server

Read and follow the instructions on Enabling the Rate Limiting Server document.

Demo Installation

If you are using the TSB demo installation, you already have rate limit service running and ready to use, and can skip this section.

If you intend to use the same rate limiting server in a multi-cluster setup, all clusters must point to the same Redis backend and domain

Deploy httpbin Service

Follow the instructions in this document to create the httpbin service. You can skip the sections "Expose the httpbin Service", "Create Certificates", and "Onboard httpbin Application".

Create TrafficSetting

Create a TrafficSetting object in a file named service-to-service-rate-limiting-traffic-setting.yaml. In this example the rate limit is set to maximum of 4 requests per minute per path. Replace the organization and tenant with appropriate values

apiVersion: traffic.tsb.tetrate.io/v2
kind: TrafficSetting
metadata:
  organization: <organization>
  tenant: <tenant>
  workspace: httpbin
  group: httpbin-traffic
  name: httpbin-traffic-settings-ratelimit
spec:
  inbound:
    rateLimiting:
      settings:
        rules:
        - dimensions:
          - header:
              name: ":path"
              value:
                prefix: "/"
          limit:
            requestsPerUnit: 4
            unit: MINUTE

Apply the manifest using tctl:

tctl apply -f service-to-service-rate-limiting-traffic-setting.yaml

Deploy sleep Service

Since you will be configuring service-to-service rate limiting, another service to act as a client to your httpbin service is necessary.

Follow the instructions in this document to create the sleep service. You can skip the section on "Create a sleep Workspace".

Testing

You can test the rate limiting by sending HTTP requests from the sleep service to httpbin service, and observe the rate limiting take effect after a certain number of requests.

To send a request from sleep service, you need to identify the pod within your sleep service. Execute the following command to find out the pod name:

kubectl get pod -n sleep -l app=sleep -o jsonpath={.items..metadata.name}

Then send a request from this pod to the httpbin service, which should be reachable at http://httpbin.httpbin:8000. Make sure to replace the value for sleep-pod with an appropriate value:

kubectl exec <sleep-pod> -n sleep -c sleep -- \
  curl http://httpbin.httpbin:8000/get \
    -s \
    -o /dev/null \
    -w "%{http_code}\n" \
    -H "X-B3-Sampled: 1"

Repeat executing the above command more than 4 times. After 4 requests, the response code that you see should change from 200 to 429.

Since the rate limiting rule was based on the request path, accessing another path on the httpbin, you should see a 200 response again:

kubectl exec <sleep-pod> -n sleep -c sleep -- \
  curl http://httpbin.httpbin:8000/headers \
    -s \
    -o /dev/null \
    -w "%{http_code}\n" \
    -H "X-B3-Sampled: 1"

Similar to the previous example, repeating the above command more than 4 times should result in the rate limiting activating, and you should start getting a 429 instead of 200.