Super dumb question: on Datomic Cloud, do system names have to be globally unique? By “globally” I mean unique even across completely different AWS accounts. Context: I have a stack failing to create and the only thing that seems to be a potential source of confusion is that I’m recycling a system name I had used in a different AWS account.

do you know where it's failing? something like a collision on the S3 bucket name might be an issue

@luchini: we have datomic systems in different accounts with the exact same name, so I think the answer is no, but if you try to recreate one in the same account with the same name, it's a bit more work if you don't want to reuse the existing storage

@luchini I’ve had stack creations fail at various times with brand new names as well.

@eoliphant this is what we got "The following resource(s) failed to create: [CodeDeployApplication, LoadBalancer, DatomicCodeBucket, HostedZone, BastionInstanceProfile]."

Thanks @stijn and @joe.lane… I’ll try a few more times. Thanks for disproving my theory 😄

@luchini there are some resources that don’t get destroyed when you delete a stack. you’ll want to look in the tag editor to search for anything from that system name to delete it explicitly

https://docs.datomic.com/cloud/operation/monitoring.html#tags

Thanks @marshall

We automate the creation and deletion of Datomic Cloud stacks. As you probably know, deleting a Datomic stack does not delete all the resources that the stack created. You need to follow these steps to entirely delete the stack https://docs.datomic.com/cloud/operation/deleting.html#deleting-storage. We wrote this clj script to automate the deletion of all the resources a Datomic Cloud system creates and thought the community may find it useful. https://gist.github.com/kennyjwilli/55007211070a260044c8e6abcb54dd5b.

I think I also had to delete an IAM policy (datomic-admin-datomic-eu-central-1), which isn't mentioned in the docs @marshall

Are there any recommendations around modeling schemas for dealing with many-to-many relationships? I’m trying to match up information to a main entity from three different data sources, and each of the integrations has their own id for the main entity. However, there is some ambiguity in the matching such one id from a given data provider can be many ids from another provider. I’m trying to determine whether it’s better to model as refs with a cardinality of many on the main entity, or if I should create ‘linkage’ records that are effectively tuples of the 3 ids that might go together. At the end of the day, I have to pare down the results so that the main entity has exactly one reference to the data from each of the other integrations, by doing a ‘best match’ with some code logic. All of that is admittedly a bit abstract, so I guess my question boils down to whether there are any recommendations for doing many-to-many relationships among more than two entities using refs or values.