Absolutely start with logging and metrics to pinpoint the problem area. Also consider setting up alerts and monitoring to catch issues early. Worst case, rolling back to a previous stable version can help stop the bleeding while you investigate further. Debugging in prod is tough but having a systematic approach and the right tooling is key.
I start by gathering detailed logs and metrics to understand the problem scope and potential causes. Then I methodically isolate variables, reproduce the issue in a test environment if possible, and use debugging tools to trace the root cause.
Replies