Skip to content

Kubernetes Deployment of two separate akka microservices which communicate via gRPC

Notifications You must be signed in to change notification settings

Michael-List/akka-microservice-sample

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

akka-microservice-sample

This is a akka kubernetes deployment example. It runs on minikube and uses the newest akka-management release and the sbt native packager for container creation.

  • A backend service which starts a akka cluster on 3 nodes (pods) and counts the greeting count for a given name
  • A frontend service which uses akka discovery to get the gRPC endpoint of service 1 to ask the cluster

arhchitecture // draw.io

Quickstart

Steps to run the demo cluster:

  • install minikube brew cask install minikube
  • configure docker eng in terminal eval $(minikube docker-env)
  • go to k8s folder cd k8s
  • run build script ./build-images.sh
  • configure kubernetes cluster roles before first run kubectl apply -f rbac-init.yaml
  • run service 1 (backend) first kubectl apply -f service1.yaml
  • run service 2 (frontend) kubectl apply -f service2.yaml
  • get your minikube ip minikube ip
  • get the service port of your NodePort Service 2 kubectl get services
  • call you minikube ip on the given NodePort with your favorite browser - The http result should be Service 2 is ok
    • if this doesn't work use minikube service service2 which shows the URL you have to use and opens your favourite browser - The http result in the browser should be Service 2 is ok
  • call the get path greet with a name e.g http://192.168.99.100:32462/greet/innFactory - The http result should be something like Hello innFactory - The actorsystem greeted you 1 times!

Update service

Example for service1:

  • make your code changes
  • in the file service1/build.sbt update the version in Docker value to 0.2 for example
  • go to k8s folder cd k8s
  • run build script ./build-images.sh
  • rollout new version kubectl set image deployment/service1 service1=innfactory-test/service1:0.2
  • check the status of the deployment with kubectl get pods

Monitoring and Tracing

There are 3 branches with different tools for monitoring and tracing:

  • branch lightbend_monitoring: Simple Monitoring example with Lightbend Monitoring.
    • folder lightbend_monitoring in the lightbend_monitoring branch.
  • branch kamon_jaeger: Monitoring example with Kamon and tracing example with Jaeger.
    • folder monitoring-and-tracing in the kamon_jaeger branch.
  • branch istio: Example Istio integration.
    • folder istio in the istio branch.

Notes

ServiceDiscovery is made with akka discovery and a bit dirty, because it does not use the cluster ip of service 1 for the grpc calls. It just tooks the first resolution address from the pod list with selector app=service1 and blocks service2 with Await. For a production use-case this should be changed and the the grpc client binding should use the result of the kubernetes-api-discovery from future and not from Await with hard .get!!!. Logging is also very dirty and there are a lot of println.

Blogpost

Contributors

innFactory - Cloud Engineering Experts - akka, grpc, k8s, kubernetes

About

Kubernetes Deployment of two separate akka microservices which communicate via gRPC

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Scala 97.6%
  • Shell 2.4%