A question buyers ask AI

Our logs are scattered across services and nobody can find anything during incidents. What kind of observability setup should we look for to fix that?

Answers to this question are being collected.