This is a surprising take. Accounting for duplicate events is such a common pattern in distributed systems that it never occurred to me to think of it as an iss

This is a surprising take. Accounting for duplicate events is such a common pattern in distributed systems that it never occurred to me to think of it as an issue. What is your preferred approach?