On-call isn't supposed to be this hard

Tbh, I think troubleshooting / investigation is really a lot different from feature development.

Wrong fixes upstream are really bad because software downstream starts relying on these.

You have to be unafraid of deep dives, but at the same time remain not overinvested in the hypothesis you came up with. You can be utterly wrong. Many paths lead to Rome, yada yada.

It's mainly a game of eliminating possibilities. Think of it like Sudoku. And the less hints you have, the more difficult or even impossible it is.