Caching responses

Use the Gloo Edge Enterprise caching server to cache responses from upstream services for quicker response times.

This feature is available only for Gloo Edge Enterprise v1.12.x and later.

The Gloo Edge Enterprise caching filter is an extension (implementing filter) of the Envoy cache filter and takes advantage of all the cache-ability checks that are applied. However, Gloo Edge also provides the ability to store the cached objects in a Redis instance, including Redis configuration options such as setting a password.

When you enable caching during installation, the caching server deployment is automatically created for you and is managed by Gloo Edge. Then, you must configure an HTTP or HTTPS listener to cache responses for its upstream services. When the listener routes a request to an upstream, the response from the upstream is automatically cached for one hour. All subsequent requests receive the cached response.

Deploy the caching server

Create a caching server during Gloo Edge Enterprise installation time, and specify any Redis overrides.

  1. Install Gloo Edge Enterprise version 1.12.x or later by using Helm. In your values.yaml file, specify the following settings:

    • Caching server: Set global.extensions.caching.enabled: true to enable the caching server deployment.
    • Redis overrides: By default, the caching server uses the Redis instance that is deployed with Gloo Edge. To use your own Redis instance, such as in production deployments:
      • Set redis.disabled to true to disable the default Redis instance.
      • Set to the name of the Redis service instance. If the instace is an external service, set the endpoint of the external service as the value.
      • For other Redis override settings, see the Redis section of the Enterprise Helm chart values.
  2. Verify that the caching server is deployed.

    kubectl --namespace gloo-system get all | grep caching

    Example output:

    pod/caching-service-5d7f867cdc-bhmqp                       1/1     Running   0          74s
    service/caching-service                       ClusterIP   <none>          8085/TCP                                               77s
    deployment.apps/caching-service                       1/1     1            1           77s
    replicaset.apps/caching-service-5d7f867cdc                       1         1         1       76s
  3. Optional: You can also check the debug logs to verify that caching is enabled.

    glooctl debug logs

    Search the output for caching to verify that you have log entries similar to the following:

    {"level":"info","ts":"2022-08-02T20:47:30.057Z","caller":"radix/server.go:31","msg":"Starting our basic redis caching service","version":"1.12.0"}
    {"level":"info","ts":"2022-08-02T20:47:30.057Z","caller":"radix/server.go:35","msg":"Created redis pool for caching","version":"1.12.0"}

Configure caching for a listener

After you enable the caching server, you must configure caching on a listener defined in a Gateway CRD.

  1. Edit the Gateway CRD where your listener is defined.

    kubectl edit gateway -n gloo-system gateway-proxy
  2. Specify the caching server in the httpGateway.options section. Currently, all paths for all upstreams that are served by a listener are cached.

       kind: Gateway
         name: gateway-proxy
         namespace: gloo-system
         bindAddress: ::
         bindPort: 8080
         - gateway-proxy
                 name: caching-service
                 namespace: gloo-system