mirror of
https://github.com/jlengrand/helidon.git
synced 2026-03-10 08:21:17 +00:00
3adfa0d190
* MetricSupport and RegistryFactory can be accessed in any order now. Signed-off-by: Tomas Langer <tomas.langer@oracle.com> * Metrics refactoring continued - using a singleton RegistryFactory and removed singleton from MetricsSupport. Delayed initialization of registries in MetricsSupport to latest possible moment. Signed-off-by: Tomas Langer <tomas.langer@oracle.com> * Checkstyle fixes. Signed-off-by: Tomas Langer <tomas.langer@oracle.com> * Fixes in statically initialized singleton. Signed-off-by: Tomas Langer <tomas.langer@oracle.com>
TODOs Demo Application
If you want to run behind a proxy, you need to configure the following in application.yaml of both services (find appropriate existing google-login provider configuration):
providers:
- google-login:
proxy-host: "proxy.host"
client-id: "1048216952820-6a6ke9vrbjlhngbc0al0dkj9qs9tqbk2.apps.googleusercontent.com"
Build and start the applications:
mvn clean install
mvn -f demo-frontend/pom.xml generate-resources docker:build
mvn -f demo-backend/pom.xml generate-resources docker:build
mvn -f demo-backend/cassandra/pom.xml generate-resources docker:build
and then
docker run -d --name zipkin -p 9411:9411 openzipkin/zipkin
docker run -d --name helidon-todos-cassandra -p 9042:9042 --link zipkin helidon.demos/io/helidon/demo/helidon-todos-cassandra
docker run -d --name helidon-todos-backend -p 8854:8854 --link zipkin --link helidon-todos-cassandra helidon.demos/io/helidon/demo/helidon-todos-backend
docker run -d --name helidon-todos-frontend -p 8080:8080 --link zipkin --link helidon-todos-backend helidon.demos/io/helidon/demo/helidon-todos-frontend
or
docker-compose up
Link map:
| URL | Description |
|---|---|
| http://localhost:8080/index.html | Main page of the demo |
| http://localhost:8080/metrics | Prometheus metrics (frontend metrics) |
| http://localhost:8080/env | Environment name (from configuration) |
| http://localhost:9411/zipkin/ | Tracing page for Zipkin (served from docker started above) |
Kubernetes setup: This assumes that the cluster is running DNS for services and pods. This is required for pods to find services by name as shown in the configuration.
export KUBECONFIG=./path/to/admin.conf
# From application root
cd k8s
# Create or update deployments and services for demo
kubectl apply -f k8s-deployment.yml
It takes a few minutes for all containers to start when running in Kubernetes. After all containers are started, use these URLs:
| URL | Description |
|---|---|
| http://localhost:30080/index.html | Main page of the demo |
| http://localhost:30080/metrics | Prometheus metrics (frontend metrics) |
| http://localhost:30080/env | Environment name (from configuration) |
| http://localhost:30011/zipkin | Tracing page for Zipkin (served from docker started above) |