zrlog 0.2.0

Logging with Zirconium-based configuration and supports audit logging

Zirconium Logging (ZrLog)

This package adds logging support using the Zirconium configuration tool and TOML, with an extension for supporting logging audit events.

Defining Logging Parameters

Configuration for the logging module can be added in TOML under the logging key. The entries correspond to those supported by logging.config.dictConfig() with a few additions to support audit logging. For example:

# .logging.toml (or your own app configuration file that you've registered) 
[logging]
version = 1

[logging.root]
level = "INFO"
handlers = ["console"]

[logging.handlers.console]
class = "logging.StreamHandler"
formatter = "brief"
level = "WARNING"
stream = "ext://sys.stdout"

[logging.formatters.brief]
format = "%(message)s [%(levelname)s]"

Of note, if you want to change a specific logger (which often have dots in the name), you must quote the key:

[logging.loggers."module.foo.bar"]
level = "WARNING"

Setting up logging

Logging can be initialized at the appropriate place in your code. This should be AFTER you have registered all your configuration with Zirconium but before you want to do any logging.

import zrlog
zrlog.init_logging()

For test cases, configuration is often not done until the test case level, but you can use the no_config parameter to tell ZrLog to ignore the configuration.

import zrlog
zrlog.init_logging(no_config=True)

Additional Logging Levels

This package adds three additional levels of logging:

• audit(), which is intended to be used only with the Python auditing system as described below. The logging level is set to 1.
• trace(), which is intended to be even more detailed than debug(). The logging level is set to 5.
• out(), which is ranked between INFO and WARNING and is intended to be used to log user output for command-line applications or key events that need to be tracked for auditing purposes. The logging level is set to 25.

These are configured as methods on the getLogger() class as well as on logging itself for the root logger. Putting these together with existing levels, we recommend the following usages:

Level	Meaning
critical	An error so severe has occurred that the system may now crash.
error	An error has occurred and something that was expected to happen did not happen.
warning	Something unexpected happen or a problem may occur in the future.
out	Something expected has happened that needs to be tracked in a production environment (e.g. user login).
info	Something expected has happened that does not need tracking but can be useful to confirm normal operation.
debug	Additional detail for debugging an issue
trace	Even more detail for debugging an issue
audit	Python auditing output only

Logging Audit Events

This package provides a system for turning sys.audit() events into log records using a thread-based queue. This is necessary because audit events don't play nicely with the logging subsystem, leading to inconsistent errors if the logger log() method is called directly from the audit hook. Audit logging must be enabled specifically by setting the with_audit flag:

# .logging.toml
[logging]
with_audit = true

While the default level is "AUDIT", you can change this to any of the logging level prefixes by specifying the audit_level:

# .logging.toml
[logging]
with_audit = true
audit_level = "INFO"

One specific event can cause further problems: sys._getframe() is called repeatedly from the logging subsystem in Python (in 3.8 at least). These audit events are NOT logged by default, but logging of them can be enabled by turning off the omit_logging_frames flag.

# .logging.toml
[logging]
with_audit = true
omit_logging_frames = false

Audit events are logged (by default) at the AUDIT level which is below TRACE; your logger and handler must be set to that level to see these events:

[logging.root]
level = "AUDIT"
handlers = ["console"]

[logging.handlers.console]
class = "logging.StreamHandler"
formatter = "brief"
level = "AUDIT"
stream = "ext://sys.stdout"

Change Log

version 0.2.0

• Updated the audit handling thread to use a threading.Event to end itself rather than a boolean flag.
• Added the no_config flag, mostly to be used in test suites to ensure everything still works properly.
• Several documentation cleanup items