Stanley Brinkerhoff wrote: > I have a rough service application template [1] that I am using for > multiple projects. These projects all log to a directory using a date > based logger that is running in a dedicated thread with a queue. This > allows multithreaded services to properly log and not retain multiple > file handles to the log file. > > Every couple days the log rollover fails. When the logging fails the > current log file is no longer present, and no future log messages are > retained.
How is the rollover done? Close, Rename, Reopen? Is the log stored on a network share? There can be unexpected latencies when doing file system operations across a network share. You might consider eliminating a rename, so that you just close the current log file and open a new one with the date embedded in the name. -- Tim Roberts, t...@probo.com Providenza & Boekelheide, Inc. _______________________________________________ python-win32 mailing list python-win32@python.org https://mail.python.org/mailman/listinfo/python-win32