The logger is still not multiprocess capable! This needs to be fixed.

I added comments with more information.

bin/linspector

@@ -33,7 +33,7 @@ from linspector.core.monitors import Monitors
 
 # i currently only increase the 3rd number because the goal is that 0.19.* will become the first
 # stable version.
-__version__ = '0.19.56.dev1'
+__version__ = '0.19.57.dev1'
 __author__ = 'Johannes Findeisen <you@hanez.org>'
 
 

etc/linspector.conf

@@ -7,7 +7,7 @@ log_level= debug
 log_file = ~/code/linspector/linspector/log/linspector.log
 ; number of log files to be kept. i recommend to use the lowest sensible value for keeping performance high. so set
 ; the size of the log file higher to increase the log history.
-log_file_count = 10
+log_file_count = 20
 ; log file size in megabytes as int. the default is 10000000 bytes (10MiB) set in the code if not configured here.
 log_file_size = 1
 ; log file size in bytes as int. you can set to bytes if you want to set a value lower then 1 megabyte. but it will

linspector/core/logger.py

@@ -13,6 +13,11 @@ from logging import handlers
 logger = getLogger('linspector')
 
 
+# wow, this is not process capable! i thought in a last commit that the problem ist fixed, but it
+# isn't!!! see here for more details:
+# https://docs.python.org/3/howto/logging-cookbook.html#logging-to-a-single-file-from-multiple-processes
+# a different approach on handling this can be to use a exclusive log file for each process but
+# this is not what i want.
 # maybe subclass it from logging.Logger...?
 class Log:
     def __init__(self, configuration, stdout, verbose):
@@ -102,6 +107,7 @@ class Log:
             import inspect
             import multiprocessing
             import threading
+            # https://www.geeksforgeeks.org/how-to-get-the-process-id-from-python-multiprocess/
             current_process = multiprocessing.current_process()
             from_stack = inspect.stack()[1]
             function_name = from_stack.function