Batch upsert using reactive programming

Do you get the same timeout if you do the operations with the blocking API?

Might be worth checking with Reactor Blockhound (BlockHound: how it works) that you’re not accidentally tying up all the reactor threads through doing blocking operations?