Service to Service Call Demonstration

A sample demonstrating service to service call patterns in a variety of scenarios:

1. Kubernetes
2. Istio (via Anthos service mesh) with services residing in a single cluster
3. Istio (via Anthos service mesh) with services residing in different clusters
4. Cloud Run

The "Producer" changes behavior based on the payload sent to it:

1. A configurable delay can be added to the response
2. A different response code can be added (say 5XX to simulate a server error)

Running it locally

Start a service application using the following command:

./gradlew -p sample-producer bootRun

Start a client to the service:

./gradlew -p sample-caller bootRun

Testing

A call to the Producer application:

curl -v -X "POST" "http://localhost:8080/producer/messages" \
     -H "Accept: application/json" \
     -H "Content-Type: application/json" \
     -d $'{
  "id": "1",
  "payload": "test",
  "delay": "1000",
  "responseCode": 200
}'

A call to the Caller, which in-turn dispatches to the Producer:

curl -X "POST" "http://localhost:8081/caller/messages" \
     -H "Accept: application/json" \
     -H "Content-Type: application/json" \
     -d $'{
  "id": "1",
  "payload": "test",
  "delay": "1000",
  "responseCode": 200
}'

A UI is available at http://localhost:8081 to test out the flow locally