'once and only once' anything is hard, including connection oriented designs, also sync calls can fail in quite similar ways as async calls as per FLP/CAP the

'once and only once' anything is hard, including connection oriented designs, also sync calls can fail in quite similar ways as async calls as per FLP/CAP the truth is that sync calls (in normal aws + k8s example) have like 50 queues between the user's kernel and your program actually doing the work, just considering the listen(2) queues, and the network card queues, and the reality is that every one of them can just drop packets on the floor whenever it wants so in reality, network programming is hard, and even sync things are just a collection of many async pieces from the network card to the userspace