Frontend log instrumentation
Context¶
Right now we only have a remote logger implementation that pushes logs to the backend. We want to also log stacktraces but without source-maps it would be hard for those stacktraces to be actually useful/readable.
Decision¶
Since our logger implementation provides an extra
parameter for additional info to add to the log entry, we will:
- make an effort to log clear and effective messages that are actually useful when read by a developer/technician
- continuously look for ways to improve our existing logging calls when needed
- optionally leverage the
extra
parameter that to provide contextual information about where the log entry is being logged from
Consequences¶
We get the ability to have effective logs available for inspection in case it's needed.