Why can't I change the level of python logger? - python

I am trying to change the loglevel of a python logger. But it does not seem to work:
import logging
#50 CRITICAL
#40 ERROR
#30 WARNING <-- default
#20 INFO
#10 DEBUG
# 0 NOTSET
logger = logging.getLogger('a')
logger.setLevel(logging.DEBUG)
print('log level', logger.getEffectiveLevel())
logger.debug('Debug') # 10
logger.info('Info') # 20
logger.warning('Warning') # 30
logger.error('Error') # 40
The output is:
log level 10
Warning
Error
But should be:
log level 10
Debug
Info
Warning
Error
Why is that and what did I do wrong?

The problem is that the default logging handler handles only the default logging levels.
You have to add a logging handler that does the actual logging.
If you need just console logging, you could use:
logger = logging.getLogger('a')
logger.addHandler(logging.StreamHandler())
logger.setLevel(logging.DEBUG)
...
This gets you:
log level 10
Debug
Info
Warning
Error
logging.basicConfig() adds a handler with some basic formatting, but usually you will define your own format for logging.

One way to solve this is to set the basic config before logging the messages. Like this:
logging.basicConfig()

Although you've set the level of the logger, the message has to pass through a handler which has a log level also. The rule is that the most restrictive log level wins. There are several solutions. Broadly you can use basicConfig to setup the logging root and it will be inherited by child loggers like "a". Or set a handler for "a".
import logging
#50 CRITICAL
#40 ERROR
#30 WARNING <-- default
#20 INFO
#10 DEBUG
# 0 NOTSET
# option - set a basic config for root logger and its descendents
#logging.basicConfig(level=logging.DEBUG)
logger = logging.getLogger('a')
logger.setLevel(logging.DEBUG)
# option = create a handler for this logger
handler = logging.StreamHandler()
handler.setLevel(logging.DEBUG)
logger.addHandler(handler)
print('log level', logger.getEffectiveLevel())
logger.debug('Debug') # 10
logger.info('Info') # 20
logger.warning('Warning') # 30
logger.error('Error') # 40

Related

logging.getLogger('name').setLevel() does not seem to work [duplicate]

The below code is copied from the documentation. I am supposed to be able to see all the info logs. But I don't. I am only able to see the warn and above even though I've set setLevel to INFO.
Why is this happening? foo.py:
import logging
logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)
logger.debug('debug message')
logger.info('info message')
logger.warn('warn message')
logger.error('error message')
logger.critical('critical message')
Output:
workingDirectory$ python foo.py
warn message
error message
critical message
Where did the info and debug messages go??
Replace the line
logger.setLevel(logging.DEBUG)
with
logging.basicConfig(level=logging.DEBUG, format='%(message)s')
and it should work as expected. If you don't configure logging with any handlers (as in your post - you only configure a level for your logger, but no handlers anywhere), you'll get an internal handler "of last resort" which is set to output just the message (with no other formatting) at the WARNING level.
Try running logging.basicConfig() in there. Of note, I see you mention INFO, but use DEBUG. As written, it should show all five messages. Swap out DEBUG with INFO, and you should see four messages.
import logging
logging.basicConfig()
logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)
logger.debug('debug message')
logger.info('info message')
logger.warn('warn message')
logger.error('error message')
logger.critical('critical message')
edit: Do you have logging set up elsewhere in your code already? Can't reproduce the exact behavior you note with the specific code provided.
As pointed by some users, using:
logging.basicConfig(level=logging.DEBUG, format='%(message)s')
like written in the accepted answer is not a good option because it sets the log level for the root logger, so it may lead to unexpected behaviours (eg. third party libraries may start to log debug messages if you set loglevel=logging.DEBUG)
In my opinion the best solution is to set log level just for your logger, like this:
import logging
logger = logging.getLogger('MyLogger')
handler = logging.StreamHandler()
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
handler.setFormatter(formatter)
logger.addHandler(handler)
logger.setLevel(logging.DEBUG)
Not really intuitive solution, but is necessary if you want to set log level only for 'MyLogger' and leave the root logger untouched.
So, why is logging.basicConfig(level=logging.DEBUG, format='%(message)s') setting the log level globally?
Well, actually it doesn't. As said, it's just changing the configuration of the root logger and, as described in the python documentation:
Loggers should NEVER be instantiated directly, but always through the
module-level function logging.getLogger(name). Multiple calls to
getLogger() with the same name will always return a reference to the
same Logger object.
So, logging.basicConfig is creating a StreamHandler with a default Formatter and adding it to the root logger.
The point is that if any other library is using the "root logger", you're going to set that log level for that library too so it can happen that you start to see debug logs from third party libraries.
This is why I think it's better to create your own logger and set your own formatters and handlers, so you can leave the root logger untouched.
This is technically also an "answer", because it can "solve" the problem. BUT I definitely DO NOT like it. It is not intuitive, and I lost 2+ hours over it.
Before:
import logging
logger = logging.getLogger('foo')
logger.setLevel(logging.INFO)
logger.info('You can not see me')
# Or you can just use the following one-liner in command line.
# $ python -c "import logging; logger = logging.getLogger('foo'); logger.setLevel(logging.INFO); logger.info('You can not see me')"
After:
import logging
logging.debug('invisible magic') # <-- magic
logger = logging.getLogger('foo')
logger.setLevel(logging.INFO)
logger.info('But now you can see me')
# Or you can just use the following one-liner in command line.
$ python -c "import logging; logging.debug('invisible magic'); logger = logging.getLogger('foo'); logger.setLevel(logging.INFO); logger.info('But now you see me')"
PS: Comparing it to the current chosen answer, and #Vinay-Sajip's explanation, I can kind of understand why. But still, I wish it was not working that way.
If you want this to work WITHOUT basicConfig, you have to first set up the lowest possible level you'll log onto the logger. Since the logger sets a minimum threshold, handlers which have a lower threshold but belong to the same logger won't get those lower threshold messages since they're ignored by the logger in the first place. Intuitive, but not obvious.
We start by doing this:
lgr = logging.getLogger(name)
lgr.setLevel(logging.DEBUG)
Then, set up the handlers with the different levels you need, in my case I want DEBUG logging on stdout and INFO logging to a rotating file, so I do the following:
rot_hndlr = RotatingFileHandler('filename.log',
maxBytes=log_size,
backupCount=3)
rot_hndlr.setFormatter(formatter)
rot_hndlr.setLevel(logging.INFO)
lgr.addHandler(rot_hndlr)
stream_hndlr = logging.StreamHandler()
stream_hndlr.setFormatter(stream_formatter)
lgr.addHandler(stream_hndlr)
Then, to test, I do this:
lgr.debug("Hello")
lgr.info("There")
My stdout (console) will look like this:
Hello
There
and my filename.log file will look like this:
There
In short, change the level in logging.basicConfig will influence the global settings.
You should better set level for each logger and the specific handler in the logger.
The following is an example that displays all levels on the console and only records messages >= errors in log_file.log. Notice the level for each handler is different.
import logging
# Define logger
logger = logging.getLogger('test')
# Set level for logger
logger.setLevel(logging.DEBUG)
# Define the handler and formatter for console logging
consoleHandler = logging.StreamHandler() # Define StreamHandler
consoleHandler.setLevel(logging.DEBUG) # Set level
concolsFormatter = logging.Formatter('%(name)s - %(levelname)s - %(message)s') # Define formatter
consoleHandler.setFormatter(concolsFormatter) # Set formatter
logger.addHandler(consoleHandler) # Add handler to logger
# Define the handler and formatter for file logging
log_file = 'log_file'
fileHandler = logging.FileHandler(f'{log_file}.log') # Define FileHandler
fileHandler.setLevel(logging.ERROR) # Set level
fileFormatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s') # Define formatter
fileHandler.setFormatter(fileFormatter) # Set formatter
logger.addHandler(fileHandler) # Add handler to logger
# Test
logger.debug('This is a debug')
logger.info('This is an info')
logger.warning('This is a warning')
logger.error('This is an error')
logger.critical('This is a critical')
Console output:
# Test
test - DEBUG - This is a debug
test - INFO - This is an info
test - WARNING - This is a warning
test - ERROR - This is an error
test - CRITICAL - This is a critical
File log_file.log content:
2021-09-22 12:50:50,938 - test - ERROR - This is an error
2021-09-22 12:50:50,938 - test - CRITICAL - This is a critical
To review your logger's level:
logger.level
The result should be one of the following:
10 # DEBUG
20 # INFO
30 # WARNING
40 # ERROR
50 # CRITICAL
To review your handlers's levels:
logger.handlers
[<StreamHandler stderr (DEBUG)>,
<FileHandler ***/log_file.log (ERROR)>]
The accepted answer does not work for me on Win10, Python 3.7.2.
My solution:
logging.basicConfig(level=logging.DEBUG)
logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)
It's order sensitive.
You have to set the basicConfig of the root logger to DEBUG, then you can set the level of your individual loggers to more restrictive levels.
This is not what I expected. Here is what I had to do:
#!/usr/bin/env python3
import logging
# by default this is WARNING. Leaving it as WARNING here overrides
# whatever setLevel-ing you do later so it seems they are ignored.
logging.basicConfig(level=logging.DEBUG)
l = logging.getLogger(__name__)
l.setLevel(level=logging.INFO)
# if I hadn't called basicConfig with DEBUG level earlier,
# info messages would STILL not be shown despite calling
# setLevel above. However now debug messages will not be shown
# for l because setLevel set it to INFO
l.warning('A warning message will be displayed')
l.info('A friendly info message will be displayed')
l.debug('A friendly debug message will not be displayed')
Most of the answers that I've found for this issue uses the basicConfig of the root logger.
It's not helpful for those who intend to use multiple independent loggers that were not initialised with basicConfig. The use of basicConfig implies that the loglevels of ALL loggers will be changed. It also had the unfortunate side effect of generating duplicate logs.
So I tried over several days experimenting with different ways to manipulate the loglevels and came up with one that finally worked.
The trick was to not only change the log levels of all the handlers but also the all the handlers of the parent of the logger.
def setLevel(self, infoLevel):
# To dynamically reset the loglevel, you need to also change the parent levels as well as all handlers!
self.logger.parent.setLevel(infoLevel)
for handler in self.logger.parent.handlers:
handler.setLevel(infoLevel)
self.logger.setLevel(infoLevel)
for handler in self.logger.handlers:
handler.setLevel(infoLevel)
The inspiration came from the fact that the basicConfig changes the root logger settings, so I was trying to do the same without using basicConfig.
For those that are interested, I did a little Python project on Github that illustrates the different issues with setting loglevel of the logger (it works partially), proves the SLogger (Sample Logger) implementation works, and also illustrates the duplicate log issue with basicConfig when using multiple loggers not initialised with it.
https://github.com/FrancisChung/python-logging-playground
TLDR: If you're only interested in a working sample code for the logger, the implentation is listed below
import logging
CRITICAL = 50
FATAL = CRITICAL
ERROR = 40
WARNING = 30
WARN = WARNING
INFO = 20
DEBUG = 10
NOTSET = 0
class SLogger():
"""
SLogger : Sample Logger class using the standard Python logging Library
Parameters:
name : Name of the Logger
infoLevel : logging level of the Logger (e.g. logging.DEBUG/INFO/WARNING/ERROR)
"""
def __init__(self, name: str, infoLevel=logging.INFO):
try:
if name is None:
raise ValueError("Name argument not specified")
logformat = '%(asctime)s %(levelname)s [%(name)s %(funcName)s] %(message)s'
self.logformat = logformat
self.name = name.upper()
self.logger = logging.getLogger(self.name)
self.logger.setLevel(infoLevel)
self.add_consolehandler(infoLevel, logformat)
except Exception as e:
if self.logger:
self.logger.error(str(e))
def error(self, message):
self.logger.error(message)
def info(self, message):
self.logger.info(message)
def warning(self, message):
self.logger.warning(message)
def debug(self, message):
self.logger.debug(message)
def critical(self, message):
self.logger.critical(message)
def setLevel(self, infoLevel):
# To dynamically reset the loglevel, you need to also change the parent levels as well as all handlers!
self.logger.parent.setLevel(infoLevel)
for handler in self.logger.parent.handlers:
handler.setLevel(infoLevel)
self.logger.setLevel(infoLevel)
for handler in self.logger.handlers:
handler.setLevel(infoLevel)
return self.logger.level
def add_consolehandler(self, infoLevel=logging.INFO,
logformat='%(asctime)s %(levelname)s [%(name)s %(funcName)s] %(message)s'):
sh = logging.StreamHandler()
sh.setLevel(infoLevel)
formatter = logging.Formatter(logformat)
sh.setFormatter(formatter)
self.logger.addHandler(sh)
Create object the right way, e.g. inspired by Google:
import logging
formatter = logging.Formatter('%(asctime)s %(threadName)s: %(message)s')
log = logging.getLogger(__name__)
log.setLevel(logging.DEBUG)
handler = logging.StreamHandler()
handler.setFormatter(formatter)
log.addHandler(handler)
log.debug('debug message')
log.info('info message')
log.warn('warn message')
log.error('error message')
log.critical('critical message')
2022-11-22 23:17:59,342 MainThread: debug message
2022-11-22 23:17:59,342 MainThread: info message
2022-11-22 23:17:59,342 MainThread: warn message
2022-11-22 23:17:59,342 MainThread: error message
2022-11-22 23:17:59,342 MainThread: critical message
As pointed out by #ManuelFedele, logging.basicConfig is not a good solution.
#VinaySajip explained that the setLevel is ignored because the logger is using the internal handler "of last resort", whose level is set to WARNING.
This explanation was also helpful:
The Handler.setLevel() method, just as in logger objects, specifies the lowest severity that will be dispatched to the appropriate destination. Why are there two setLevel() methods? The level set in the logger determines which severity of messages it will pass to its handlers. The level set in each handler determines which messages that handler will send on.
So a good solution is to add a handler to the logger, with the appropriate level:
import logging
logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)
ch = logging.StreamHandler()
ch.setLevel(logging.DEBUG) # or whatever level should be displayed on the console
logger.addHandler(ch)
Output
>>> logger.debug('debug message')
debug message
>>> logger.info('info message')
info message
>>> logger.warn('warn message')
<stdin>:1: DeprecationWarning: The 'warn' method is deprecated, use 'warning' instead
warn message
>>> logger.error('error message')
error message
>>> logger.critical('critical message')
critical message

Why this python code filter out logs lower than INFO?

import logging
# logging.basicConfig(level=logging.DEBUG)
consoleHandler = logging.StreamHandler()
consoleHandler.setLevel(logging.DEBUG)
consoleHandler.setFormatter(logging.Formatter("%(asctime)s; %(levelname)s; %(message)s"))
logging.getLogger().addHandler(consoleHandler)
logging.debug('debug')
logging.info('test')
logging.warn('warn')
logging.error('error')
logging.fatal('fatal')
I got only the last three logs, which are equal to or higher than WARN. Could anyone please tell me why?
You set the handler's level, not the logger's. The root logger's level is still set to the root's default of WARNING, so it won't pass any messages less severe than WARNING to the handler.
As #user2357112 has already mentioned, set level for root logger
logging.getLogger().setLevel(logging.DEBUG)
Your code would be like:
import logging
# logging.basicConfig(level=logging.DEBUG)
consoleHandler = logging.StreamHandler()
consoleHandler.setLevel(logging.DEBUG)
consoleHandler.setFormatter(logging.Formatter("%(asctime)s; %(levelname)s; %(message)s"))
logging.getLogger().addHandler(consoleHandler)
logging.getLogger().setLevel(logging.DEBUG)
logging.debug('debug')
logging.info('test')
logging.warn('warn')
logging.error('error')
logging.fatal('fatal')#Output
#output
# 2017-12-18 12:15:49,647; DEBUG; debug
# 2017-12-18 12:15:49,648; INFO; test
# 2017-12-18 12:15:49,648; WARNING; warn
# 2017-12-18 12:15:49,648; ERROR; error
# 2017-12-18 12:15:49,648; CRITICAL; fatal

Not able to log to two different places using Python logging module

Forgive me, still learning how to use the logging module. So, I have two handlers, one to log everything to a file and one specifically to log the message when triggered into Hipchat.
I don't quite understand what it means to have root vs non-root loggers. As it is below, it will write what I want to my log file AND displays both messages into the hipchat room. So I found another stack post that says to create two multiple "non root loggers." So I tried it with logger = logging.getLogger('test1') and hipchat_logger = logging.getLogger('test2') but there's no message in Hipchat (it does write to the log file).
logger = logging.getLogger()
hipchat_logger = logging.getLogger()
# File handler
file_handler = logging.FileHandler("mylog.log", "a")
logger.addHandler(file_handler)
logger.setLevel(logging.DEBUG)
# Hipchat handler
hipchat_handler = hiplogging.HipChatHandler(MY_TOKEN, CHATROOM)
hipchat_handler.setLevel(logging.DEBUG)
hipchat_logger.addHandler(hipchat_handler)
# Testing
logger.info("Hi, this is a test. I should not see this in Hipchat")
hipchat_logger.info("Hi, display me in Hipchat")
Problem is in the first two lines. "Multiple calls to getLogger() with the same name will always return a reference to the same Logger object.", cf. here.
So a solution would be (modified after discussion in comments):
logger = logging.getLogger()
hipchat_logger = logging.getLogger("hipchat")
EDIT
After more research on this topic: The problem is certainly not the missing root logger. There is a problem with loglevels:
Logging messages which are less severe than lvl will be ignored. When a logger is created, the level is set to NOTSET (which causes all messages to be processed when the logger is the root logger, or delegation to the parent when the logger is a non-root logger). Note that the root logger is created with level WARNING.
So what does happen here?
# implicit root logger with loglevel WARNING
# both child-logger: loglevel NOTSET
logger = logging.getLogger('file')
hipchat_logger = logging.getLogger('hipchat')
# ...create and add file handler...
# change 'file'-loglevel to DEBUG
logger.setLevel(logging.DEBUG)
# ...create hipchat handler...
# change loglevel of the handler(!) to DEBUG.
# 'hipchat'-loglevel is still NOTSET
hipchat_handler.setLevel(logging.DEBUG)
# ...add hipchat handler ...
# logger -> loglevel DEBUG -> INFO gets logged.
logger.info("Hi, this is a test. I should not see this in Hipchat")
# hipchat_logger -> loglevel NOTSET -> delegate to root-logger
# root-logger -> loglevel WARNING -> INFO not logged (being below threshold)
hipchat_logger.info("Hi, display me in Hipchat")
So to solve this issue, either explicitly set the loglevel of root logger to DEBUG (as in the code above). Or simply set the loglevel of the second logger accordingly:
hipchat_logger.setLevel(logging.DEBUG)
A question that might remain is answered elsewhere: What is the point of setlevel in a handler?
One with your code problem is that calling logging.getLogger() multiple times returns the same logger.
# both of these are same logger instances
logger = logging.getLogger()
hipchat_logger = logging.getLogger()
To get two different instance, give different names to different loggers. Also, with no name parameter, you get the root logger.
root_logger = logging.getLogger()
logger = logging.getLogger("file")
hipchat_logger = logging.getLogger("hipchat")
#now getting the logger with same name will return same logger
this_is_same_as_logger = logging.getLogger("file")

About NOTSET in python logging

As the logger.setLevel doc says:
When a logger is created, the level is set to NOTSET (which causes all messages to be processed when the logger is the root logger, or delegation to the parent when the logger is a non-root logger). Note that the root logger is created with level WARNING.
so I think if I create a root logger, with level NOTSET, the debug and info log will display.
The code use basicConfig to set root logger's level to NOTSET is right:
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import logging
logging.basicConfig(level=logging.NOTSET)
log = logging.getLogger()
log.debug('debug')
log.info('info')
log.warning('warning')
log.error('error')
log.critical('critical')
and output is:
DEBUG:root:debug
INFO:root:info
WARNING:root:warning
ERROR:root:error
CRITICAL:root:critical
But if I create a root logger, and add handler with NOTSET level to it, such as:
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import logging
log = logging.getLogger()
hd = logging.StreamHandler()
hd.setLevel(logging.NOTSET)
log.addHandler(hd)
log.debug('debug')
log.info('info')
log.warning('warning')
log.error('error')
log.critical('critical')
the output is:
warning
error
critical
but I think it will also output the debug and info message.
OK, I misunderstand the Logger's level and Handler's Level, in the doc:
The setLevel() method, just as in logger objects, specifies the lowest severity that will be dispatched to the appropriate destination. Why are there two setLevel() methods? The level set in the logger determines which severity of messages it will pass to its handlers. The level set in each handler determines which messages that handler will send on.
If I change code to this, will be ok:
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import logging
log = logging.getLogger()
log.setLevel(logging.NOTSET) # Set Logger's level to NOTSET, default is WARNING
#print "Logger's Level: ", log.level
hd = logging.StreamHandler()
hd.setLevel(logging.NOTSET)
#print "Handler's Level: ", hd.level
log.addHandler(hd)
log.debug('debug')
log.info('info')
log.warning('warning')
log.error('error')
log.critical('critical')
I also thought this was confusing. I interpreted NOTSET as don't produce logs. But...
The level and handlers entries are interpreted as for the root logger, except that if a non-root logger’s level is specified as NOTSET, the system consults loggers higher up the hierarchy to determine the effective level of the logger. source
and
The level is interpreted as for loggers, and NOTSET is taken to mean ‘log everything’. source
This makes sense because NOTSET (0) which is below DEBUG (10)
Let's explain by example that UNSET does NOT do anything when creating concrete instances of loggers (for some mysterious reason):
def logging_unset_level():
"""My sample logger explaining UNSET level
Resources:
- https://stackoverflow.com/questions/21494468/about-notset-in-python-logging
- https://www.youtube.com/watch?v=jxmzY9soFXg&t=468s
- https://github.com/CoreyMSchafer/code_snippets/tree/master/Logging-Advanced
"""
import logging
logger = logging.getLogger(__name__) # loggers are created in hierarchy using dot notation, thus __name__ ensures no name collisions.
print(f'DEFAULT VALUE: logger.level = {logger.level}')
file_handler = logging.FileHandler(filename='my_log.log')
log_format = "{asctime}:{levelname}:{lineno}:{name}:{message}" # see for logrecord attributes https://docs.python.org/3/library/logging.html#logrecord-attributes
formatter = logging.Formatter(fmt=log_format, style='{')
file_handler.setFormatter(fmt=formatter)
stdout_stream_handler = logging.StreamHandler(stream=sys.stdout)
stdout_stream_handler.setLevel(logging.INFO)
log_format = "{name}:{levelname}:-> {message}" # see for logrecord attributes https://docs.python.org/3/library/logging.html#logrecord-attributes
formatter = logging.Formatter(fmt=log_format, style='{')
stdout_stream_handler.setFormatter(fmt=formatter)
logger.addHandler(hdlr=file_handler)
logger.addHandler(hdlr=stdout_stream_handler)
logger.log(logging.NOTSET, 'notset')
logger.debug('debug')
logger.info('info')
logger.warning('warning')
logger.error('error')
logger.critical('critical')
the output to screen is:
DEFAULT VALUE: logger.level = 0
__main__:WARNING:-> warning
__main__:ERROR:-> error
__main__:CRITICAL:-> critical
the output to stdout is:
2020-04-15 17:00:38,384:WARNING:200:__main__:warning
2020-04-15 17:00:38,384:ERROR:201:__main__:error
2020-04-15 17:00:38,384:CRITICAL:202:__main__:critical
as expected from the question's body, when the logger is instantiated its setting is UNSET. If you are using the basic config for some reason it does work as the docs imply it should everything from UNSET to higher get logged BUT when instantiating loggers concretely this setting makes only error messages work for some reason (find me a reference!).
Ok, now if you run it again but increase the level of the top logger everything works as you would expect (except the unset):
logger.setLevel(logging.DEBUG)
The output of stdout (note the stdout handler has level INFO in my example for pedagogical reasons, different from the file handler, trying to show inheritance of the .level value from the top logger):
DEFAULT VALUE: logger.level = 10
__main__:INFO:-> info
__main__:WARNING:-> warning
__main__:ERROR:-> error
__main__:CRITICAL:-> critical
and the output of the logger file my_log.log (this one DOES output the debug statements! :D ):
2020-04-15 17:05:58,782:DEBUG:198:__main__:debug
2020-04-15 17:05:58,784:INFO:199:__main__:info
2020-04-15 17:05:58,784:WARNING:200:__main__:warning
2020-04-15 17:05:58,784:ERROR:201:__main__:error
2020-04-15 17:05:58,784:CRITICAL:202:__main__:critica
is as expected. It inherited the top level loggers level so it printed everything from DEBUG level upwards.

log messages appearing twice with Python Logging

I'm using Python logging, and for some reason, all of my messages are appearing twice.
I have a module to configure logging:
# BUG: It's outputting logging messages twice - not sure why - it's not the propagate setting.
def configure_logging(self, logging_file):
self.logger = logging.getLogger("my_logger")
self.logger.setLevel(logging.DEBUG)
self.logger.propagate = 0
# Format for our loglines
formatter = logging.Formatter("%(asctime)s - %(name)s - %(levelname)s - %(message)s")
# Setup console logging
ch = logging.StreamHandler()
ch.setLevel(logging.DEBUG)
ch.setFormatter(formatter)
self.logger.addHandler(ch)
# Setup file logging as well
fh = logging.FileHandler(LOG_FILENAME)
fh.setLevel(logging.DEBUG)
fh.setFormatter(formatter)
self.logger.addHandler(fh)
Later on, I call this method to configure logging:
if __name__ == '__main__':
tom = Boy()
tom.configure_logging(LOG_FILENAME)
tom.buy_ham()
And then within say, the buy_ham module, I'd call:
self.logger.info('Successfully able to write to %s' % path)
And for some reason, all the messages are appearing twice. I commented out one of the stream handlers, still the same thing. Bit of a weird one, not sure why this is happening. I'm assuming I've missed something obvious.
You are calling configure_logging twice (maybe in the __init__ method of Boy) : getLogger will return the same object, but addHandler does not check if a similar handler has already been added to the logger.
Try tracing calls to that method and eliminating one of these. Or set up a flag logging_initialized initialized to False in the __init__ method of Boy and change configure_logging to do nothing if logging_initialized is True, and to set it to True after you've initialized the logger.
If your program creates several Boy instances, you'll have to change the way you do things with a global configure_logging function adding the handlers, and the Boy.configure_logging method only initializing the self.logger attribute.
Another way of solving this is by checking the handlers attribute of your logger:
logger = logging.getLogger('my_logger')
if not logger.handlers:
# create the handlers and call logger.addHandler(logging_handler)
If you are seeing this problem and you're not adding the handler twice then see abarnert's answer here
From the docs:
Note: If you attach a handler to a logger and one or more of its
ancestors, it may emit the same record multiple times. In general, you
should not need to attach a handler to more than one logger - if you
just attach it to the appropriate logger which is highest in the
logger hierarchy, then it will see all events logged by all descendant
loggers, provided that their propagate setting is left set to True. A
common scenario is to attach handlers only to the root logger, and to
let propagation take care of the rest.
So, if you want a custom handler on "test", and you don't want its messages also going to the root handler, the answer is simple: turn off its propagate flag:
logger.propagate = False
I'm a python newbie, but this seemed to work for me (Python 2.7)
while logger.handlers:
logger.handlers.pop()
The handler is added each time you call from outside. Try Removeing the Handler after you finish your job:
self.logger.removeHandler(ch)
In my case I'd to set logger.propagate = False to prevent double printing.
In below code if you remove logger.propagate = False then you will see double printing.
import logging
from typing import Optional
_logger: Optional[logging.Logger] = None
def get_logger() -> logging.Logger:
global _logger
if _logger is None:
raise RuntimeError('get_logger call made before logger was setup!')
return _logger
def set_logger(name:str, level=logging.DEBUG) -> None:
global _logger
if _logger is not None:
raise RuntimeError('_logger is already setup!')
_logger = logging.getLogger(name)
_logger.handlers.clear()
_logger.setLevel(level)
ch = logging.StreamHandler()
ch.setLevel(level)
# warnings.filterwarnings("ignore", "(Possibly )?corrupt EXIF data", UserWarning)
ch.setFormatter(_get_formatter())
_logger.addHandler(ch)
_logger.propagate = False # otherwise root logger prints things again
def _get_formatter() -> logging.Formatter:
return logging.Formatter(
'[%(asctime)s] [%(name)s] [%(levelname)s] %(message)s')
This can also happen if you are trying to create a logging object from the parent file. For e.g.
This is the main application file test.py
import logging
# create logger
logger = logging.getLogger('simple_example')
logger.setLevel(logging.DEBUG)
# create console handler and set level to debug
ch = logging.StreamHandler()
ch.setLevel(logging.DEBUG)
# create formatter
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
# add formatter to ch
ch.setFormatter(formatter)
# add ch to logger
logger.addHandler(ch)
def my_code():
# 'application' code
logger.debug('debug message')
logger.info('info message')
logger.warning('warn message')
logger.error('error message')
logger.critical('critical message')
And below is the parent file main.py
import test
test.my_code()
The output of this will print only once
2021-09-26 11:10:20,514 - simple_example - DEBUG - debug message
2021-09-26 11:10:20,514 - simple_example - INFO - info message
2021-09-26 11:10:20,514 - simple_example - WARNING - warn message
2021-09-26 11:10:20,514 - simple_example - ERROR - error message
2021-09-26 11:10:20,514 - simple_example - CRITICAL - critical message
But if we had a parent logging object, then it will be printed twice. For e.g. if this is the parent file
import test
import logging
logging.basicConfig(level=logging.DEBUG,format='%(asctime)s - %(name)s - %(levelname)s - %(message)s')
test.my_code()
The the output will be
2021-09-26 11:16:28,679 - simple_example - DEBUG - debug message
2021-09-26 11:16:28,679 - simple_example - DEBUG - debug message
2021-09-26 11:16:28,679 - simple_example - INFO - info message
2021-09-26 11:16:28,679 - simple_example - INFO - info message
2021-09-26 11:16:28,679 - simple_example - WARNING - warn message
2021-09-26 11:16:28,679 - simple_example - WARNING - warn message
2021-09-26 11:16:28,679 - simple_example - ERROR - error message
2021-09-26 11:16:28,679 - simple_example - ERROR - error message
2021-09-26 11:16:28,679 - simple_example - CRITICAL - critical message
2021-09-26 11:16:28,679 - simple_example - CRITICAL - critical message
A call to logging.debug() calls logging.basicConfig() if there are no root handlers installed. That was happening for me in a test framework where I couldn't control the order that test cases fired. My initialization code was installing the second one. The default uses logging.BASIC_FORMAT that I didn't want.
It seems that if you output something to the logger (accidentally) then configure it, it is too late. For example, in my code I had
logging.warning("look out)"
...
ch = logging.StreamHandler(sys.stdout)
root = logging.getLogger()
root.addHandler(ch)
root.info("hello")
I would get something like (ignoring the format options)
look out
hello
hello
and everything was written to stdout twice. I believe this is because the first call to logging.warning creates a new handler automatically, and then I explicitly added another handler. The problem went away when I removed the accidental first logging.warning call.
I was struggling with the same issue in the context of multiple processes. (For the code see the docs which I was following almost verbatim.) Namely, all log messages originating from any of the child processes got duplicated.
My mistake was to call worker_configurer(),
def worker_configurer(logging_queue):
queue_handler = logging.handlers.QueueHandler(logging_queue)
root = logging.getLogger()
root.addHandler(queue_handler)
root.setLevel(level)
both in the child processes and also in the main process (since I wanted the main process to log stuff, too). The reason this led to trouble (on my Linux machine) is that on Linux the child processes got started through forking and therefore inherited the existing log handlers from the main process. That is, on Linux the QueueHandler got registered twice.
Now, preventing the QueueHandler from getting registered twice in the worker_configurer() function is not as trivial as it seems:
Logger objects like the root logger root have a handlers property but it is undocumented.
In my experience, testing whether any([handler is queue_handler for handler in root.handlers]) (identity) or any([handler == queue_handler for handler in root.handlers]) (equality) fails after forking, even if root.handlers seemingly contains the same QueueHandler. (Obviously, the previous two expressions can be abbreviated by queue_handler in root.handlers, since the in operator checks for both identity and equality in the case of lists.)
The root logger gets modified by packages like pytest, so root.handlers and root.hasHandlers() are not very reliable to begin with. (They are global state, after all.)
The clean solution, therefore, is to replace forking with spawning to prevent these kinds of multiprocessing bugs right from the start (provided you can live with the additional memory footprint, of course). Or to use an alternative to the logging package that doesn't rely on global state and instead requires you to do proper dependency injection but I'm digressing… :)
With that being said, I ended up going for a rather trivial check:
def worker_configurer(logging_queue):
queue_handler = logging.handlers.QueueHandler(logging_queue)
root = logging.getLogger()
for handler in root.handlers:
if isinstance(handler, logging.handlers.QueueHandler):
return
root.addHandler(queue_handler)
root.setLevel(level)
Obviously, this will have nasty side effects the second I decide to register a second queue handler somewhere else.
From the docs:
"Loggers have the following attributes and methods. Note that Loggers should NEVER be instantiated directly, but always through the module-level function logging.getLogger(name).
Multiple calls to getLogger() with the same name will always return a reference to the same Logger object".
Make sure you don't initialise your loggers with the same name
I advise you to initialise the logger with __name__ as name param i.e:
import logging
logger = logging.getLogger(__name__)
NOTE:
even if you init a loggers from other modules with same name, you will still get the same logger, therefore calling i.e logger.info('somthing') will log as many times as you initiated the logger class.
I was getting a strange situation where console logs were doubled but my file logs were not. After a ton of digging I figured it out.
Please be aware that third party packages can register loggers. This is something to watch out for (and in some cases can't be prevented). In many cases third party code checks to see if there are any existing root logger handlers; and if there isn't--they register a new console handler.
My solution to this was to register my console logger at the root level:
rootLogger = logging.getLogger() # note no text passed in--that's how we grab the root logger
if not rootLogger.handlers:
ch = logging.StreamHandler()
ch.setLevel(logging.INFO)
ch.setFormatter(logging.Formatter('%(process)s|%(levelname)s] %(message)s'))
rootLogger.addHandler(ch)
If you are using any config for logging, For instance log.conf
In .conf file you can do it by adding this line in the [logger_myLogger] section: propagate=0
[logger_myLogger]
level=DEBUG
handlers=validate,consoleHandler
qualname=VALIDATOR
propagate=0
I had the same issue. In my case, it was not due to handlers or duplicate initial configuration but a stupid typo. In main.py I was using a logger object but in my_tool.py I was directly calling to the logging module by mistake, hence after invoking functions from my_tool module everything was messed up and the messages appeared duplicated.
This was the code:
main.py
import logging
import my_tool
logger_name = "cli"
logger = logging.getLogger(logger_name)
logger.info("potato")
logger.debug("potato)
my_tool.function()
logger.info("tomato")
my_tool.py
import logging
logger_name = "cli"
logger = logging.getLogger(logger_name)
# some code
logging.info("carrot")
and the result
terminal
>> potato
>> potato
>> carrot
>> tomato
>> tomato
Adding to others' useful responses...
In case you are NOT accidentally configuring more than one handler on your logger:
When your logger has ancestors(root logger is always one) and they have their own handlers, they will also output when your logger outputs (by default), which will create duplicates. You have two options:
Don't propagate your log event to your ancestors by setting:
my_logger.propagate = False
If you only have one ancestor(root logger), you could directly configure their handler instead. For example:
# directly change the formatting of root's handler
root_logger = logging.getLogger()
roots_handler = root_logger.handlers[0]
roots_handler.setFormatter(logging.Formatter(': %(message)s')) # change format
my_logger = logging.getLogger('my_logger') # my_logger will use new formatting
If you use the standard construction logger = logging.getLogger('mymodule') and then accidentally mistype loggger as logging i.e.
logger = logging.getLogger('mymodule')
# configure your handlers
logger.info("my info message") # won't make duplicate
logging.info("my info message") # will make duplicate logs
then this will cause duplicate messages to come up because the call to logging creates a new logger.

Categories