Django Logging Demystified with Lee Trout - DjangoCon US 2022

Python Testing

Django logging demystified: learn how to harness the power of structured logging, where, when, and how; from log record construction to log file rotation, and beyond.

Key takeaways

Structured logging is key to demystifying Django logging. It’s about adding explicit context to log messages, making them easier to understand.
Python’s logging module is essential to understanding Django logging. It’s a complex system with many components, including loggers, handlers, filters, and formatters.
Loggers are hierarchical, with the root logger at the top. Each logger can have a level, filters, and handlers.
Handlers control where log messages are sent, such as a file, console, or email.
Filters determine whether log messages are processed by handlers.
Formatters control the format of log messages.
Levels (CRITICAL, ERROR, WARNING, INFO, DEBUG) determine the severity of log messages.
Effective logging requires understanding how all these components interact.
Structured logging can be achieved using typed dictionaries, data classes, or serialized log records.
Django comes with a default logging configuration, but it’s not inheriting settings from other apps.
The make_record helper can be used to create a log record.
Loggers can be bound to a specific context, making it easier to log related events.
Rotation and filtering are important concepts in logging configuration.
The logging module’s extra keyword can be used to add custom data to log records.
Logging is essential for testing and debugging, as it allows you to inspect and understand system behavior.
Effective logging also requires a philosophy of logging, including considering logging as a form of asynchronous debugging.
Log records can be used to generate metrics and tracing information.
Structured logging enables more advanced features like log aggregation and anomaly detection.
Python’s logging module has many built-in features and tools to help with logging configuration.
The logging.config module can be used to configure logging programmatically.
Rotating log files is an important consideration in logging configuration.
The logging.handlers module provides various handlers, such as RotatingFileHandler and SMTPHandler.
Django’s server formatter can be customized to include additional information in log messages.
Effective logging requires understanding of how logging interacts with other components, such as the ORM and signals.
Structured logging can be used to generate log entries in a specific format.
PyLint can be used to lint log messages and enforce logging best practices.
APM (Application Performance Monitoring) tools can be used to instrument and observe application behavior.
Structured logging can be used to generate metrics and tracing information.
Effective logging requires a culture of logging, including making it a priority and considering it a form of testing.

Django Logging Demystified with Lee Trout - DjangoCon US 2022

More talks