Spring Cloud Netflix Turbine is widely used to aggregates multiple Hystrix metrics streams so that it can be monitored in a single view using Hystrix dashboard. In this tutorial, we will demonstrate how to use them on Azure Spring Cloud.
Note
Netflix Hystrix is widely used in many existing Spring Cloud apps but it is no longer in active development. If you are developing new project, you can consider using resilience4j instead. Circuit Breaker Dashboard for resilience4j dependents on micrometer, which is currently not supported for Azure Spring Cloud, thus will not be covered by this tutorial.
Clone the sample repository to your develop environment. It is forked from this repository, check out the original blog if you want to learn more.
git clone https://github.com/yucwan/spring-cloud.git
cd spring-cloud/spring-turbine
Build the 3 applications that will be used in this tutorial.
- user-service: A simple REST service that has a single endpoint of
/personalized/{id} - recommendation-service: A simple REST service that has a single endpoint of
/recommendations, which will be called by user-service. - hystrix-turbine: A Hystrix dashboard service to display Hystrix streams and a Turbine service aggregating Hystrix metrics stream from other services.
mvn clean package -D skipTests -f user-service/pom.xml
mvn clean package -D skipTests -f recommendation-service/pom.xml
mvn clean package -D skipTests -f hystrix-turbine/pom.xml
Be aware that our apps are not using Config Server, thus no need to config Config Server for Azure Spring Cloud before deployment.
az spring-cloud app create -n user-service --is-public
az spring-cloud app create -n recommendation-service
az spring-cloud app create -n hystrix-turbine --is-public
az spring-cloud app deploy -n user-service --jar-path user-service/target/user-service.jar
az spring-cloud app deploy -n recommendation-service --jar-path recommendation-service/target/recommendation-service.jar
az spring-cloud app deploy -n hystrix-turbine --jar-path hystrix-turbine/target/hystrix-turbine.jar
After all the apps are running and discovered, access user-service with the path https://yuchensp-user-service.azuremicroservices.io/personalized/1 from your browser.
You should get the following output if user-service can access recommendation-service. Otherwise please try refreshing the web page a few times.
[{"name":"Product1","description":"Description1","detailsLink":"link1"},{"name":"Product2","description":"Description2","detailsLink":"link3"},{"name":"Product3","description":"Description3","detailsLink":"link3"}]You can access hystrix-turbine with the path https://<SERVICE-NAME>-hystrix-turbine.azuremicroservices.io/hystrix from your browser, which shows the Hystrix dashboard running in this app.
Paste the Turbine stream url https://<SERVICE-NAME>-hystrix-turbine.azuremicroservices.io/turbine.stream?cluster=default and click Monitor Stream, and you will see the dashboard working. If you are not viewing anything, just hit the user-service endpoints to generate the streams.
You can now play with your Circuit Breaker Dashboard.
Note
Be aware that in production, the Hystrix dashboard and metrics stream should not be exposed to internet.
Hystrix metrics streams are also accessible from test-endpoint. As a backend service, we didn't assign a public end-point for recommendation-service, let's show its metrics with test-endpoint at https://primary:<KEY>@<SERVICE-NAME>.test.azuremicroservices.io/recommendation-service/default/actuator/hystrix.stream.
As a web app, Hystrix dashboard should also be working on test endpoint. However, it is not working properly for two reasons: First, using test endpoint will change the base URL from / to /<APP-NAME>/<DEPLOYMENT-NAME>. Second, the web app is using absolute path for static resource. To get it worked on test endpoint, you might need to manually edit the <base>in the front-end files.