A simple DMN service to evaluate a loan approval and generate tracing events that might be consumed by the Trusty service.
You will need:
The default configuration pushes the tracing events to the kafka topic kogito-tracing-decision and the DMN models used by the kogito application to kogito-tracing-model under the group-id kogito-runtimes. Edit the application.properties file to change the configuration. The property names are the following:
kogito.addon.tracing.decision.kafka.bootstrapAddress: The address used in the initial connection to find a bootstrap server on the cluster of n brokers (no default value, this is mandatory).kogito.addon.tracing.decision.kafka.topic.name : The topic name for the decision tracing events (default kogito-tracing-decision).kogito.addon.tracing.decision.kafka.topic.partitions : How many partitions to use for the decision tracing events topic (default 1).kogito.addon.tracing.decision.kafka.topic.replicationFactor : The replication factor of the data for the decision tracing events topic (default 1).kogito.addon.tracing.decision.asyncEnabled: Use an asynchronous callback with the results of the send (success or failure) instead of waiting for the Future to complete (default true).kogito.addon.tracing.model.kafka.topic.name : The topic name for the DMN models used by the kogito application (default kogito-tracing-model).mvn clean compile spring-boot:run
mvn clean package java -jar target/dmn-tracing-springboot.jar
or on Windows
mvn clean package java -jar target\dmn-tracing-springboot.jar
Once the service is up and running, you can use the following example to interact with the service.
Returns penalty information from the given inputs -- driver and violation:
Given inputs:
{ "Bribe": 0, "Client": { "age": 0, "existing payments": 0, "salary": 0 }, "Loan": { "duration": 0, "installment": 0 }, "SupremeDirector": "yes" }
Curl command (using the JSON object above):
curl -X POST -H 'Accept: application/json' -H 'Content-Type: application/json' -d '{"Bribe": 0,"Client": {"age": 0,"existing payments": 0,"salary": 0},"Loan": {"duration": 0,"installment": 0},"SupremeDirector": "yes"}' http://localhost:8080/LoanEligibility
or on Windows:
curl -X POST -H "Accept: application/json" -H "Content-Type: application/json" -d "{\"Bribe\": 0,\"Client\": {\"age\": 0,\"existing payments\": 0,\"salary\": 0},\"Loan\": {\"duration\": 0,\"installment\": 0},\"SupremeDirector\": \"yes\"}" http://localhost:8080/LoanEligibility
As response, penalty information is returned.
Example response:
{ "Eligibility": "No", "Judgement": null, "Loan": { "duration": 0, "installment": 0 }, "SupremeDirector": "yes", "Bribe": 0, "Client": { "existing payments": 0, "salary": 0, "age": 0 }, "Is Enough?": 0, "Decide": null }
When the tracing addon is enabled, the tracing events are emitted and pushed to a Kafka broker. The Trusty Service can consume such events and store them on a storage. The Trusty Service exposes then some api to consume the information that has been collected. A docker-compose example is provided in the current folder. In particular, when docker-compose up is run, a Kafka broker, an Infinispan container and the nightly build of the trusty service are deployed. Once the services are up and running, after a decision has been evaluated, you can access the trusty service swagger at localhost:8081/swagger-ui.html and try to query what are the evaluations of the last day at localhost:8081/v1/executions for example.