It’s weird. We’ve started seeing some "Duplicate Entry for key PRIMARY” errors that looks to revolve around a couple of EntitySave commands. We’re not currently using transactions - just been letting the end of request do the flushing for us - and I don’t get the feeling that that’s best common practice.
We’ve temporarily taken out the section of code that appears to be the problem, but I’m curious how you folks tend to implement your transactions? I’m guessing that most entry-point type service calls are just encapsulated with the transaction tag? I’ll do some poking around, but all suggestions/leads welcomed.
Thanks, and thanks for the book tip.