Vinay Sajip <vinay_sa...@yahoo.co.uk> added the comment: I'm unable to reproduce this problem with Python 2.5.2 on either Windows XP or Ubuntu Hardy.
I used a test script (httpd.py, attached). This sets up an HTTP server which you can use to trigger logging events. I created a time simulator to allow testing in a timely way. This monkey-patches the binding for the time module in logging and logging.handlers to return either the real time or the simulated time. Once you have started the server (just run the script), you can trigger a logging event by running wget -o /dev/null http://localhost:9022/ and you can initiate simulated timing by running wget -o /dev/null http://localhost:9022/s The simulated time is set to the next midnight, to facilitate rollover. I then ran wget -o /dev/null http://localhost:9022/ three times, then wget -o /dev/null http://localhost:9022/s once, then wget -o /dev/null http://localhost:9022/ again three times. The only files created were: vi...@zeta-hardy:~$ ls -l httpd.log* -rw-r--r-- 1 vinay vinay 192 2009-06-10 16:46 httpd.log -rw-r--r-- 1 vinay vinay 256 2009-06-10 16:46 httpd.log.2009-06-10 which is as expected, and the contents are: vi...@zeta-hardy:~$ cat httpd.log.2009-06-10 2009-06-10 16:46:22,999 pid: 512 DEBUG - OK 2009-06-10 16:46:22 2009-06-10 16:46:23,765 pid: 512 DEBUG - OK 2009-06-10 16:46:23 2009-06-10 16:46:24,406 pid: 512 DEBUG - OK 2009-06-10 16:46:24 2009-06-10 16:46:25,974 pid: 512 DEBUG - OK 2009-06-10 16:46:25 and vi...@zeta-hardy:~$ cat httpd.log 2009-06-11 00:00:01,260 pid: 512 DEBUG - OK 2009-06-11 00:00:01 2009-06-11 00:00:01,923 pid: 512 DEBUG - OK 2009-06-11 00:00:01 2009-06-11 00:00:02,547 pid: 512 DEBUG - OK 2009-06-11 00:00:02 which is again as expected. Please try out the test script in your environment and feed back the results you get. ---------- resolution: -> works for me status: open -> pending Added file: http://bugs.python.org/file14258/httpd.py _______________________________________ Python tracker <rep...@bugs.python.org> <http://bugs.python.org/issue6252> _______________________________________ _______________________________________________ Python-bugs-list mailing list Unsubscribe: http://mail.python.org/mailman/options/python-bugs-list/archive%40mail-archive.com
