Enable Proxy Caching

Use the KongPlugin CRD to enable proxy caching on a KongService. The following example:

• Caches 200 OK responses
• Applies to GET and HEAD requests
• Targets responses with text/plain content type
• Stores cached data in memory
• Sets the cache TTL to 300 seconds (5 minutes)

echo '
kind: KongPlugin
apiVersion: configuration.konghq.com/v1
metadata:
  name: proxy-cache-all-endpoints
  namespace: kong
plugin: proxy-cache
config:
  response_code:
  - 200
  request_method:
  - GET
  - HEAD
  content_type:
  - text/plain; charset=utf-8
  cache_ttl: 300
  strategy: memory
' | kubectl apply -f -

Apply the plugin by annotating the KongService with its name:

kubectl annotate -n kong kongservice service konghq.com/plugins=proxy-cache-all-endpoints --overwrite

Send a request to the Route and inspect the X-Cache-Status header. The first request finds nothing in the cache, so it is sent to the upstream Service and its response is cached:

curl -i "http://$PROXY_IP/echo" \
     --no-progress-meter --fail-with-body 

X-Cache-Status: Miss

Send the same request again. This time the response is served from the cache:

curl -i "http://$PROXY_IP/echo" \
     --no-progress-meter --fail-with-body 

X-Cache-Status: Hit

Responses stay cached for 300 seconds, as set in the cache_ttl field of the plugin configuration.