Fork me on GitHub

As of the newest tools.deps.alpha "0.5.435" (TDEPS-9, you can now use Datomic Pro! May I suggest that someone on the Datomic team add a note about the below to this page?

;; ~/.m2/settings.xml
<settings xmlns=""

;; deps.edn

 {com.datomic/datomic-pro {:mvn/version "0.9.5697"}}

 {"" {:url ""}}}
thanks @alexmiller @dominicm !


Hi! What does :db.error/cycle-in-affinity mean exactly? (I can work around the issue but I’d like to understand better what’s going on)


I’m having trouble distinguishing between whether to use datomic’s api (, or the client api ( The APIs are similar, but different enough. FWIW I’m using datomic starter.


@brycecovert if you're just trying to tinker, I've found Datomic Cloud to be very quick to get to a working environment. I suppose it depends what your goals are.


I am mostly just wanting to learn and see if it makes sense to integrate datomic into an existing system. if it does, deploying it through docker would be the most straightforward way. My assumption is that would be wrapping up datomic starter


are one of these apis meant to be used for cloud and the other on-prem/starter?


Oh, I think I’ve got it. One is the peer api and the other is the client api.


My understanding is that there is both a peer and a client library available for use with On-Prem. Cloud has its own separate client library (and only a client library).


I don't have any experience with launching Datomic via Docker, but it looks like there are plenty of options over on Docker Hub. (;isOfficial=0&amp;page=1&amp;pullCount=0&amp;q=datomic&amp;starCount=0)


Those are all going to be the unlicensed variant I imagine they wouldn't be used for production.


>I am mostly just wanting to learn and see if it makes sense to integrate datomic into an existing system. I'd personally recommend launching a Datomic Cloud (Solo) instance in this case. It's easy to get set up, and it can directly transition into a production deployment without any API changes necessary.


I have one running in my AWS account that I just create lots of ephemeral databases on for testing ideas and for development


That’s helpful - I’ll try out solo.


is there any way you can use the peer library with the cloud version on AWS?


I'm mostly interested in taking advantage of the client side cache