How does the logging module work?

Q: How does the logging module work?

**Target Audience:** Python developers who use `print` for diagnostics, or who have fought duplicated or missing log lines, and want to understand how `logging` routes a message **Difficulty:** Medium **Python version:** Checked on CPython 3.11 (and in this page on Pyodide 0.25.1, which is CPython 3.11.3); 3.12 additions were checked on python3.12 > **How to read this doc:** Each idea is explained in plain language first, then tagged with `📌 Interview term:` so you also learn the exact word an interviewer expects. Every output block in the **Verified** section was copied from a real run of the examples on this page. ## 1. Why This Even Matters — A Story First Think of a large office mail room. Every department writes notes: "printer jammed", "invoice 42 failed", "server on fire". Each note is stamped with how urgent it is. The mail room then decides, per note: does anyone care about this urgency? Which inboxes does it go to — the department's own tray, the floor manager's, the head

The problem

Question presented to candidate: "Explain how Python's logging module works. What are loggers, handlers and formatters, how does a log call decide whether and where to output something, and how would you set up logging for an application versus a library?"

What a strong answer should cover:

Loggers come from logging.getLogger(__name__) and form a tree by dotted name under the root logger; each has a level, and an unset level (NOTSET) inherits the nearest ancestor's (the root defaults to WARNING)
The flow of a record: logger level check, logger filters, then every handler on that logger and, while propagate is true, on each ancestor; each handler has its own level, filters and formatter
Handlers decide the destination (StreamHandler, FileHandler, RotatingFileHandler, QueueHandler), formatters the text (%(asctime)s %(levelname)s %(name)s %(message)s)
Use %-style arguments (log.info("user %s", uid)) so formatting is skipped when the record is filtered out; log.exception() inside except adds the traceback; extra= or a LoggerAdapter add context fields
Configuration: basicConfig once at the application entry point (a no-op if the root already has handlers, unless force=True, added in 3.8), or dictConfig for real apps; libraries only call getLogger and add a NullHandler, never configure the root
Pitfalls: duplicate lines from handlers on both a child and the root, disable_existing_loggers defaulting to true in dictConfig, and the last-resort handler that prints WARNING and above to stderr when nothing is configured

Code / implementation expected: Optional — configure a handler and formatter, show a module logger inheriting a level, and one propagation or exception() example.

logging
stdlib

Try it yourself

Sketch your approach before you read the solution. That is what the interview tests.

Stuck? Get a hint.

A short nudge toward the approach. It will not give the answer away.

Solution

Target Audience: Python developers who use print for diagnostics, or who have fought duplicated or missing log lines, and want to understand how logging routes a message Difficulty: Medium Python version: Checked on CPython 3.11 (and in this page on Pyodide 0.25.1, which is CPython 3.11.3); 3.12 additions were checked on python3.12

How to read this doc: Each idea is explained in plain language first, then tagged with 📌 Interview term: so you also learn the exact word an interviewer expects. Every output block in the Verified section was copied from a real run of the examples on this page.

1. Why This Even Matters — A Story First

Think of a large office mail room. Every department writes notes: "printer jammed", "invoice 42 failed", "server on fire". Each note is stamped with how urgent it is. The mail room then decides, per note: does anyone care about

The solution is hidden so you can try first

12 min read · 12 sections

Run the code

Edit the code and run it here. Each console line with an expected value becomes a test.

PythonLoggers, levels, handlers, formatters, exception() and propagation

Practice

PythonbasicConfig, the last-resort handler, NullHandler, dictConfig, filters and context