On Software: Logging

Posted 2023-08-07

Quite a while ago, I was thinking about settling on a steady logging convention, after always having slightly varying versions in my different projects.

First of all, I collected some reference material, and focused on defining a set of log levels (for the severity of an entry).
And then… the topic dozed off — until now.

Starting Point: Syslog

Syslog is an old and established ~~convention~~, standard (RFC 5424).
But for my liking and needs, it’s a bit too fine-grained:

Code	Severity	Description
0	Emergency	System is unusable
1	Alert	Action must be taken immediately
2	Critical	Critical conditions
3	Error	Error conditions
4	Warning	Warning conditions
5	Notice	Normal, but significant condition
6	Informational	Informational messages
7	Debug	Debug-level messages

My current convention

While investigating a few other logging systems, I discovered that most of them used (or are based on) a trimmed down list of these Syslog severity levels, sometimes with more or less diverging names and definitions.

So I compared them and reduced it to the most common ones, ran it through my own internal checker, and et voilà:
Based on my findings and preferences, I came up with this condensed version as my new convention, for now:

Code	Severity	Description
0	Critical	Indication that a crucial/core functionality is no longer working. Unrecoverable, maybe even fatal failure.
1	Error	Operation of the software is impaired, because a functionality ist not working (correctly).
2	Warning	Indication of a potential issue, or that something unexpected happened; but the software can continue to work normally.
3	Information	Notification that things are working as expected; can be ignored during normal operations.
4	Verbose	Detailed information, typically only of interest when tracing and diagnosing issues, troubleshooting (‘debugging’), and during development.

Feedback: comment@saoe.net · Categories: Development, On Software · Tags: EN

◀ Mini-Reviews #43

Mini-Reviews #44 ▶