[api-umbrella] mongodb data persistence

[robjuz]

api-umbrella comes with a build in mongodb. The chart is missing a persistence volume to mitigate data lost from the db.

It would be also nice to have the mongodb as a dependent Chart. This would enable independent scalability of the main deployment and the database.

---

I'm ready to help with the implementation :)

[wistefan]

In a cloud environment(and in containers in general), I would not recommend running multiple services inside the same container. Even though umbrella always runs multiple processes, I definitely do not recommend to enable the mongo-db(and also elasticsearch) as a service inside the container and therefore would rather not put an pvc for them in there. This would also require to make the chart a statefulset to enable horizontal scaling and further complicates the configuration.

Adding mongo(and elasticsearch) as a dependent chart(even thought that I don't think running databases productive inside kubernetes is a good idea) would be a nice addition. A contribution would be very welcome, I will probably not be able to do it myself before end of august.