From the docs,
The correct use of 'uncaughtException' is to perform synchronous cleanup of allocated resources (e.g. file descriptors, handles, etc) before shutting down the process. It is not safe to resume normal operation after 'uncaughtException'.
This makes sense to me. Performing any work whilst the application is in an unstable state does seem like asking for further trouble
I'm curious if there were any drawbacks to using a logging service, such as Winston, within the event handler function for both of the above events and if there are any ways to prevent those drawbacks (e.g., if Winston was the cause of the uncaught exception, checking that Winston wasn't the cause (I'm no sure if that's possible, btw, just an example!) before proceeding to logging)
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…