coherence-operator

Access Coherence via a Separate Proxy Tier

This sample shows how to deploy two tiers, a storage-enabled data tier and a storage-disabled proxy tier. This is a common scenario when you are using Coherence*Extend to connect to a cluster and when you want to separate the proxy tier from the data tier.

This is achieved by using two helm install commands, both of which include a sidecar container for the data tier and proxy tier cache configuration.

Return to Coherence*Extend samples / Return to Coherence Deployments samples / Return to samples

Sample files

• src/main/docker/Dockerfile - Dockerfile for creating sidecar image from which the configuration will be read at pod startup

• src/main/resources/client-cache-config.xml - Cache configuration for the extend client

• src/main/resources/conf/proxy-cache-config.xml - Cache configuration for proxy-tier

• src/main/resources/conf/storage-cache-config.xml - Cache configuration for storage-tier

Prerequisites

Ensure that you have installed the Coherence Operator by following the instructions here.

Installation Steps

1. Change to the samples/coherence-deployments/extend/proxy-tier directory. Ensure that you have your maven build environment set for JDK8, and build the project.

   $ mvn clean install -P docker
   

   This builds the Docker image with the cache configuration files and compiled Java classes, with the name in the format proxy-tier-sample:${version}. For example,

    proxy-tier-sample:1.0.0-SNAPSHOT
   

   Note: If you are running against a remote Kubernetes cluster, you must push the above image to your repository accessible to that cluster. You must also prefix the image name in your helm command as shown below.

2. Install the Coherence cluster.

   $ helm install \
      --namespace sample-coherence-ns \
      --name storage \
      --set clusterSize=3 \
      --set cluster=proxy-tier-cluster \
      --set imagePullSecrets=sample-coherence-secret \
      --set store.cacheConfig=storage-cache-config.xml \
      --set logCaptureEnabled=false \
      --set userArtifacts.image=proxy-tier-sample:1.0.0-SNAPSHOT \
      coherence/coherence
   

Once the installation is complete, run the following command to retrieve the list of pods:

   $ kubectl get pods -n sample-coherence-ns

All the three storage-coherence pods should be running and ready, as shown in the output:

   NAME                   READY   STATUS    RESTARTS   AGE
   storage-coherence-0    1/1     Running   0          4m
   storage-coherence-1    1/1     Running   0          2m   
   storage-coherence-2    1/1     Running   0          2m

1. Install the storage-disabled proxy-tier.

   Set the following properties to ensure that this release connects to the Coherence clusterSize created by the storage release:

   • --set cluster=proxy-tier-cluster - Uses the same cluster name

   • --set store.wka=storage-coherence-headless - Ensures that it can contact the cluster

   • --set cluster=proxy-tier-cluster - Ensures that the cluster name is the same

   • --set store.storageEnabled=false - Sets storage to false

   • --set store.cacheConfig=proxy-cache-config.xml - Uses proxy cache configuration from sidecar

   Note: For the proxy-tier, we are using a clusterSize of one, to save resources. You can scale out the proxy-tier for high availability purposes.

   $ helm install \
     --namespace sample-coherence-ns \
     --set cluster=proxy-tier-cluster \
     --set clusterSize=1 \
     --set store.storageEnabled=false \
     --set store.wka=storage-coherence-headless \
     --name proxy-tier \
     --set imagePullSecrets=sample-coherence-secret \
     --set store.cacheConfig=proxy-cache-config.xml \
     --set userArtifacts.image=proxy-tier-sample:1.0.0-SNAPSHOT \
     coherence/coherence
   

   To confirm that the proxy-tier has joined the cluster, you can look at the logs using:

   $ kubectl logs proxy-tier-coherence-0 -n sample-coherence-ns | grep ActualMemberSet
   

   This should return the following, which indicates that there are now four members:

   ActualMemberSet=MemberSet(Size=4
   

   You should now see three charts installed:

   $ helm ls
   NAME              	REVISION	UPDATED                 	STATUS  	CHART                            	APP VERSION   	NAMESPACE          
   coherence-operator	1       	Wed Mar 20 14:12:31 2019	DEPLOYED	coherence-operator-1.0.0-SNAPSHOT	1.0.0-SNAPSHOT	sample-coherence-ns
   proxy-tier        	1       	Wed Mar 20 14:54:57 2019	DEPLOYED	coherence-1.0.0-SNAPSHOT         	1.0.0-SNAPSHOT	sample-coherence-ns
   storage           	1       	Wed Mar 20 14:53:58 2019	DEPLOYED	coherence-1.0.0-SNAPSHOT         	1.0.0-SNAPSHOT	sample-coherence-ns
   

2. Port forward the proxy port on the proxy-tier.

   $ kubectl port-forward -n sample-coherence-ns proxy-tier-coherence-0 20000:20000
   

3. Connect via CohQL and run the commands:

   $ mvn exec:java
   

   Run the following CohQL commands to insert data into the cluster.

   insert into 'test' key('key-1') value('value-1');
   
   select key(), value() from 'test';
   Results
   ["key-1", "value-1"]
   
   select count() from 'test';
   Results
   1
   

Uninstall the Charts

Run the following command to delete both the charts installed in this sample:

$ helm delete storage proxy-tier --purge

Before starting another sample, ensure that all pods are removed from the previous sample. To remove coherence-operator, use the helm delete command.

This site is open source. Improve this page.