I never really understood how people are using a debugger these days, especially in a heavy I/O bound app such as Node.js services. I mean, once it hits the DB,

I never really understood how people are using a debugger these days, especially in a heavy I/O bound app such as Node.js services. I mean, once it hits the DB, there is no way of “stepping back”. I used to use debuggers back when I was developing desktop apps on Java using Eclipse. It made sense to use a debugger back then cause the whole app was running locally inside the same execution stack, including its state and its business logic. You were in control of how the data flow and in which direction. But nowadays, you have an API request with a short lifecycle which 90% includes handing the execution to other processes through the network. What is the purpose of using a debugger? You can just write an automated test that would automatically run on every file change.