Logging

logger level recommendations

use debug only for detail debugging information, e.g. when the data is big or the frequency is high

def complex_algorithm(items):
    for i, item in enumerate(items):
        # do some complex algorithm computation
        logger.debug('%s iteration, item=%s', i, item)

use info levels for routines, e.g. when handling requests or server state changes

def handle_request(request):
    logger.info('Handling request %s', request)
    # handle request here
    result = 'result'
    logger.info('Return result: %s', result)

def start_service():
    logger.info('Starting service at port %s ...', port)
    service.start()
    logger.info('Service is started')

use warning when it's important, but not an error. E.g. for authentication or performance warnings

def authenticate(user_name, password, ip_address):
    if user_name != USER_NAME and password != PASSWORD:
        logger.warn('Login attempt to %s from IP %s', user_name, ip_address)
        return False
    # do authentication here

use error level when something is wrong, e.g. when throwing exceptions, IO failures, no connection

def get_user_by_id(user_id):
    user = db.read_user(user_id)
    if user is None:
        logger.error('Cannot find user with user_id=%s', user_id)
        return user
    return user

there is also the critical level, but this is seldomly used

logger name recommendations

It's recommended to set the logger name to __name__, this way the configurations are shared when used in a module

exceptions

It can be helpfull to record tracebacks to see where something went wrong, as logging itself won't be enough help for troubleshooting
You should capture exceptions and record them with traceback
You can call logger methods with the exc_info=True parameter to dump traceback

try:
    open('/path/to/does/not/exist', 'rb')
except (SystemExit, KeyboardInterrupt):
    raise
except Exception, e:
    logger.error('Failed to open file', exc_info=True)

StefVanremoortele / python-logging