-
Feature Request
-
Resolution: Won't Fix
-
L1 - Blocker
-
None
-
None
Searching on github issues, I saw multiple discussions saying this feature will not supported because does not follows the micro-service architecture, so I will describe what I think is a valid use case.
I work in a company that uses Camunda with spring-boot starter as part of Microservices Architecture in multi-site and multi-branding products. This means that we could have one brand, for example "Nike", starting BPM processes in multiples countries at the same time, all this deployed in one node of my cluster app.
So, my node is a micro-service that needs to handle all this by itself with different data-sources (i.e Schemas or even different database clusters), and running additional nodes is not an option since no more infrastructure could be added at this time or any time sooner.
We achieved this by-passing the spring-boot instantiation mode of the Process Engine, this leads to dependency injections issues of our dependencies but we managed to inject directly using Spring context.
My point is, why using spring-boot starter should lead to a reduction of Camunda features such as multi-tenancy with multiple data-sources? From my consideration it should really do not reduce or not support the features that Camunda offers.
We were thinking on upgrade to Camunda EE, so we would be paying for features that are not supported on camunda-springboot-starter? It should be delegate to us the possibility of adopt good micro-service architecture or not, and it should not be restricted by the framework provided.
Thanks for consider my point of view.
This is the controller panel for Smart Panels app
- is related to
-
-
- Ready
-
- links to
