Before we get into databases, we have to address one of the attributes of a high-quality software system: managing logged information, including normal system activity, system errors, and debugging information. Logs give us an insight into the behavior of the system. How much traffic is it getting? If it's a website, which pages are people hitting the most? How many errors occur and of what kind? Do attacks occur? Are malformed requests being sent?

Log management is also an issue. Log rotation means regularly moving the log file out of the way, to start with a fresh log file. You should process logged data to produce reports. A high priority on screening for security vulnerabilities is a must.

The Twelve Factor application model suggests simply sending logging information to the console, and then some other software system captures that output and directs it to a logging service. Following their advice can reduce system complexity by having fewer things that can break. In a later chapter, we'll use PM2 for that purpose.

Let's first complete a tour of information logging as it stands right now in Notes.

When we used the Express Generator to initially create the Notes application, it configured an activity logging system using morgan:

const logger = require('morgan'); 
.. 
app.use(logger('dev')); 

This is what prints the requests on the Terminal window. Visit https://github.com/expressjs/morgan for more information.

Internally, Express uses the Debug package for debugging traces. You can turn these on using the DEBUG environment variable. We should try to use this package in our application code. For more information, visit https://www.npmjs.com/package/debug.

Finally, the application might generate uncaught exceptions. The uncaughtException error needs to be captured, logged, and dealt with appropriately.