Multi Engine Adaptation C7 - C8

Project Overview

The project entails a partial transition to Camunda 8, offering substantial advantages. It involves adopting a Multi-Engine architecture where both Camunda 7 (c7) and Camunda 8 (c8) coexist and intercommunicate through APIs.

Process Diagram:

Key Objectives

• Enhanced Flexibility: Utilizing the features of Camunda 8 to introduce greater flexibility into the system.

• Partial Migration: Implementing a phased migration approach to gradually transition to Camunda 8 while preserving existing functionality.

• Prioritized Implementation: Strategically prioritizing the migration based on feature availability and specific requirements.

End-to-End Visibility with Optimize

The Ingest API in Camunda Optimize (C7 Feature only) allows to:

• Enables a unified view of processes and activities across the organization.
• Supports cross-platform analytics and visualization.
• Facilitates process optimization and data-driven decision-making.

Via the Ingestion API we can send Camunda 8 Events to Optimize.

Implementation in a gRPC Client Interceptor

Camunda 8 Events are send to Optimize via a Interceptor. Following steps are happening:

1. Initialize gRPC Client Interceptor: Set up a gRPC client interceptor for custom logic in outgoing calls.

2. Capture Job Information: Capture relevant job data when a job is activated in a job worker.

3. Prepare Data for Ingestion: Aggregate essential data (e.g., job ID, process instance ID) for ingestion into Camunda Optimize. As a traceId, a variable with the name correlationKey is used. The correlationKey maps to the business key in Camunda 7.

4. Ingest Data into Camunda Optimize: Utilize the Ingest API endpoints to send a POST request with the prepared data.

These steps facilitate efficient integration of the Ingest API, enabling seamless transmission and analysis of job-related data in Camunda Optimize.

See Events in Optimize

You need to enable Event Based Processes in Optimize:

engines:
  'camunda-bpm':
    eventImportEnabled: true

eventBasedProcess:
  authorizedUserIds: ['demo']
  authorizedGroupIds: []
  eventImport:
    enabled: true

Now the ingested events are visible in Optimize

1. Create BPMN Diagram

2. Map Events

3. Create Reports

Limitations

• Only ActivateJob Events will be send to Optimize (No BpmnError, No Engine Events (Timer, Message received, ...))
• ActivatedJobs will be reported to Optimize, even if the completion fails

What you need to run the project

• Camunda License Key (including Optimize) (update in license.txt)
• Camunda EE Images (You can also use Camunda Run CE. Optimize requires an Enterprise License)

docker login registry.camunda.cloud
Username: my.username
Password: my.pw

Login Succeeded
docker pull registry.camunda.cloud/cambpm-ee/camunda-bpm-platform-ee:run-7.19.6
docker pull registry.camunda.cloud/optimize-ee/optimize:3.10.5

docker compose up

mvn spring-boot:run

Start an Instance

curl --location 'http://localhost:8080/engine-rest/process-definition/key/C7_First/start' \
--header 'Content-Type: application/json' \
--data '{
    "variables": {
        "payload": {
            "value": "Hello",
            "type": "String"
        }

    },
    "businessKey": "1235555x"
}'