This is the multi-page printable view of this section. Click here to print.

Return to the regular view of this page.

Configuration

1 - Example: Configuring a Java Microservice

1.1 - Externalizing config using MicroProfile, ConfigMaps and Secrets

In this tutorial you will learn how and why to externalize your microservice’s configuration. Specifically, you will learn how to use Kubernetes ConfigMaps and Secrets to set environment variables and then consume them using MicroProfile Config.

Before you begin

Creating Kubernetes ConfigMaps & Secrets

There are several ways to set environment variables for a Docker container in Kubernetes, including: Dockerfile, kubernetes.yml, Kubernetes ConfigMaps, and Kubernetes Secrets. In the tutorial, you will learn how to use the latter two for setting your environment variables whose values will be injected into your microservices. One of the benefits for using ConfigMaps and Secrets is that they can be re-used across multiple containers, including being assigned to different environment variables for the different containers.

ConfigMaps are API Objects that store non-confidential key-value pairs. In the Interactive Tutorial you will learn how to use a ConfigMap to store the application's name. For more information regarding ConfigMaps, you can find the documentation here.

Although Secrets are also used to store key-value pairs, they differ from ConfigMaps in that they're intended for confidential/sensitive information and are stored using Base64 encoding. This makes secrets the appropriate choice for storing such things as credentials, keys, and tokens, the former of which you'll do in the Interactive Tutorial. For more information on Secrets, you can find the documentation here.

Externalizing Config from Code

Externalized application configuration is useful because configuration usually changes depending on your environment. In order to accomplish this, we'll use Java's Contexts and Dependency Injection (CDI) and MicroProfile Config. MicroProfile Config is a feature of MicroProfile, a set of open Java technologies for developing and deploying cloud-native microservices.

CDI provides a standard dependency injection capability enabling an application to be assembled from collaborating, loosely-coupled beans. MicroProfile Config provides apps and microservices a standard way to obtain config properties from various sources, including the application, runtime, and environment. Based on the source's defined priority, the properties are automatically combined into a single set of properties that the application can access via an API. Together, CDI & MicroProfile will be used in the Interactive Tutorial to retrieve the externally provided properties from the Kubernetes ConfigMaps and Secrets and get injected into your application code.

Many open source frameworks and runtimes implement and support MicroProfile Config. Throughout the interactive tutorial, you'll be using Open Liberty, a flexible open-source Java runtime for building and running cloud-native apps and microservices. However, any MicroProfile compatible runtime could be used instead.

Objectives

  • Create a Kubernetes ConfigMap and Secret
  • Inject microservice configuration using MicroProfile Config

Example: Externalizing config using MicroProfile, ConfigMaps and Secrets

Start Interactive Tutorial

1.2 - Interactive Tutorial - Configuring a Java Microservice

To interact with the Terminal, please use the desktop/tablet version

2 - Configuring Redis using a ConfigMap

This page provides a real world example of how to configure Redis using a ConfigMap and builds upon the Configure a Pod to Use a ConfigMap task.

Objectives

  • Create a ConfigMap with Redis configuration values
  • Create a Redis Pod that mounts and uses the created ConfigMap
  • Verify that the configuration was correctly applied.

Before you begin

You need to have a Kubernetes cluster, and the kubectl command-line tool must be configured to communicate with your cluster. It is recommended to run this tutorial on a cluster with at least two nodes that are not acting as control plane hosts. If you do not already have a cluster, you can create one by using minikube or you can use one of these Kubernetes playgrounds:

To check the version, enter kubectl version.

Real World Example: Configuring Redis using a ConfigMap

Follow the steps below to configure a Redis cache using data stored in a ConfigMap.

First create a ConfigMap with an empty configuration block:

cat <<EOF >./example-redis-config.yaml
apiVersion: v1
kind: ConfigMap
metadata:
  name: example-redis-config
data:
  redis-config: ""
EOF

Apply the ConfigMap created above, along with a Redis pod manifest:

kubectl apply -f example-redis-config.yaml
kubectl apply -f https://raw.githubusercontent.com/kubernetes/website/main/content/en/examples/pods/config/redis-pod.yaml

Examine the contents of the Redis pod manifest and note the following:

  • A volume named config is created by spec.volumes[1]
  • The key and path under spec.volumes[1].items[0] exposes the redis-config key from the example-redis-config ConfigMap as a file named redis.conf on the config volume.
  • The config volume is then mounted at /redis-master by spec.containers[0].volumeMounts[1].

This has the net effect of exposing the data in data.redis-config from the example-redis-config ConfigMap above as /redis-master/redis.conf inside the Pod.

pods/config/redis-pod.yaml 
apiVersion: v1
kind: Pod
metadata:
  name: redis
spec:
  containers:
  - name: redis
    image: redis:5.0.4
    command:
      - redis-server
      - "/redis-master/redis.conf"
    env:
    - name: MASTER
      value: "true"
    ports:
    - containerPort: 6379
    resources:
      limits:
        cpu: "0.1"
    volumeMounts:
    - mountPath: /redis-master-data
      name: data
    - mountPath: /redis-master
      name: config
  volumes:
    - name: data
      emptyDir: {}
    - name: config
      configMap:
        name: example-redis-config
        items:
        - key: redis-config
          path: redis.conf

Examine the created objects:

kubectl get pod/redis configmap/example-redis-config

You should see the following output:

NAME        READY   STATUS    RESTARTS   AGE
pod/redis   1/1     Running   0          8s

NAME                             DATA   AGE
configmap/example-redis-config   1      14s

Recall that we left redis-config key in the example-redis-config ConfigMap blank:

kubectl describe configmap/example-redis-config

You should see an empty redis-config key:

Name:         example-redis-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
redis-config:

Use kubectl exec to enter the pod and run the redis-cli tool to check the current configuration:

kubectl exec -it redis -- redis-cli

Check maxmemory:

127.0.0.1:6379> CONFIG GET maxmemory

It should show the default value of 0:

1) "maxmemory"
2) "0"

Similarly, check maxmemory-policy:

127.0.0.1:6379> CONFIG GET maxmemory-policy

Which should also yield its default value of noeviction:

1) "maxmemory-policy"
2) "noeviction"

Now let's add some configuration values to the example-redis-config ConfigMap:

pods/config/example-redis-config.yaml 
apiVersion: v1
kind: ConfigMap
metadata:
  name: example-redis-config
data:
  redis-config: |
    maxmemory 2mb
    maxmemory-policy allkeys-lru

Apply the updated ConfigMap:

kubectl apply -f example-redis-config.yaml

Confirm that the ConfigMap was updated:

kubectl describe configmap/example-redis-config

You should see the configuration values we just added:

Name:         example-redis-config
Namespace:    default
Labels:       <none>
Annotations:  <none>

Data
====
redis-config:
----
maxmemory 2mb
maxmemory-policy allkeys-lru

Check the Redis Pod again using redis-cli via kubectl exec to see if the configuration was applied:

kubectl exec -it redis -- redis-cli

Check maxmemory:

127.0.0.1:6379> CONFIG GET maxmemory

It remains at the default value of 0:

1) "maxmemory"
2) "0"

Similarly, maxmemory-policy remains at the noeviction default setting:

127.0.0.1:6379> CONFIG GET maxmemory-policy

Returns:

1) "maxmemory-policy"
2) "noeviction"

The configuration values have not changed because the Pod needs to be restarted to grab updated values from associated ConfigMaps. Let's delete and recreate the Pod:

kubectl delete pod redis
kubectl apply -f https://raw.githubusercontent.com/kubernetes/website/main/content/en/examples/pods/config/redis-pod.yaml

Now re-check the configuration values one last time:

kubectl exec -it redis -- redis-cli

Check maxmemory:

127.0.0.1:6379> CONFIG GET maxmemory

It should now return the updated value of 2097152:

1) "maxmemory"
2) "2097152"

Similarly, maxmemory-policy has also been updated:

127.0.0.1:6379> CONFIG GET maxmemory-policy

It now reflects the desired value of allkeys-lru:

1) "maxmemory-policy"
2) "allkeys-lru"

Clean up your work by deleting the created resources:

kubectl delete pod/redis configmap/example-redis-config

What's next