In more than 10 years as an Android engineer, I’ve seen a ton of crashes that were impossible to reproduce or things that are behaving in a non-sense when reading stack traces.
Most of the time, I’ve left things as-is, considering it was something beyond my power. I’m pretty sure it’s the case for a lot of developers: you’ve seen a lot of crash reports which made non-sense without knowing what to do.
Me, when reading obscure stack traces
But thanks to Firebase Crashlytics (or any other crash reporting tool), we have ways to identify how crashes are happening and in which conditions. With the experience, you’ll identify if a crash is worth taking your time on investigations or not.
The crash
The more your application and your user base grow, the more issues will happen. It’s important to do prioritization work, as many issues can happen, and you may spend a lot of time investigating the cause.