Project/Environment Architecture

Hi,

I’ m a little bit confused:

  1. How I can create a Engine Server? Only deploying the flowable-admin app?
  2. How can I configure the engine in this case? Where to put the flowable.cfg file?

If I choose to use an embedded engine, the process definition will be deployed every time I start the application?

Hi,

if you deploy flowable-task engine with in memory configuration is created. All UI applications (-admin, -modeler,…) use this engine through REST api.
I would say you need to deploy idm app too.

I would prefer to set VM options to override default one e.g.
-Ddatasource.url=… -Ddatasource.driver=… -Ddatasource.username=… -Ddatasource.passwrod=…

The deployment depends on configuration

Martin

Ok,

What does it mean?

I’m trying to understand what kind of environment architecture use for production.
I think that a common engine, to deploy the different processes is better. but I can’t understand how the various configuration on the flowable.cfg.xml (for example the jpa configration etc) match with this architecture.
Maybe I’m a little bit confused, but I’ve never used neither flowable or activiti.
I studied the documentation, I made some example and I’m studying a solution to be proposed to the customer that is loath to the news because of the system team that manage the servers

I did not understand the rest of your reply. Could you clarify your question?

Martin

Ok, it’s not simple. I’ll try…

Maybe, for my organization, it will be better to have an admin console to deply the various project, so, the flowable-admin for example. I think something like a single admin and many project, is it good?

I also can’t understand how the various configuration on the flowable.cfg.xml (for example the jpa configration etc) match with this architecture.

May I need an example of a real project with a real production architecture…

Check org.flowable.spring.test.jpa.JpaTest in flowable source. May be it will help.

Martin

The choice is up to you:

  • either you deploy a ‘central’ server (with or without the REST API) where all processes are managed and executed. All the data will live in one datastore (note: there is multi-tenancy that can be used for the setup).
  • or, you embed the engine in each project and each project/engine has their own datasore
  • or, a mix and match depending on your use case.

Also note that the flowable.cfg.xml is but one way of instantiating your process engine. You can also opt to e.g. instantiate your process engine programmatically.