anybody using JDBI or HikariCP?

We evaluated Exposed and JDBI and went with JDBI.

what DB? postgres?

yep

How are you doing connection pooling? Does JDBI only make one connection?

We are at the very beginning of the project and we have not set up connection pooling. We will though.

we use hikari w/ jooq: https://github.com/trib3/leakycauldron/tree/master/db

We use HikariCP all the time, very good and stable pool implementation

Been using JDBI+Hikari in production for a while. No issues at all, if that's what you're worried about. 🙂

also using jdbi+hikari

How do you model your DAOs when joins are involved? eg if I have a customer that can have many orders - and I have a customer dao and an order dao — how do I get the Joined relation?

I tend to just use IDs between two separate aggregate roots. So no real navigation within the domain. If you have like an order with an address I do all of that in the order dao.

@Anders Sveen so the order dao will query the address table?

Yeah. Or depending on the structure, just a join so we get a single query. If I have to make decisions based on two aggregate roots I put that in a service, or better yet I do order.smoethingMethod(customer) .