fd leakage with run_queue

Hello,
It is quite often that the system load on a machine goes up so sendmail rejects connections.... so when run_queue runs, there are quite a few messages rejected in one process, meaning lots of logs... and then I get messages like this:
Logging error: <StampedLogger to '/var/lib/mailman/logs/error'> Traceback (innermost last): File "/usr/lib/mailman/Mailman/Logging/Logger.py", line 55, in __get_f IOError: [Errno 24] Too many open files: '/var/lib/mailman/logs/error' Original log message: [Errno 24] Too many open files: '/var/lib/mailman/logs/error' Feb 10 13:44:59 2000 processQueue: Processing of queue file /var/lib/mailman/data/mm_q.28769.1 failed: Feb 10 13:44:59 2000 processQueue: exceptions.IOError / [Errno 24] Too many open files: '/var/lib/mailman/data/mm_q.28769.1' Logging error: <StampedLogger to '/var/lib/mailman/logs/error'> Traceback (innermost last): File "/usr/lib/mailman/Mailman/Logging/Logger.py", line 55, in __get_f IOError: [Errno 24] Too many open files: '/var/lib/mailman/logs/error' Original log message: [Errno 24] Too many open files: '/var/lib/mailman/logs/error' Feb 10 13:45:00 2000 processQueue: Processing of queue file /var/lib/mailman/data/mm_q.29101.1 failed: Feb 10 13:45:00 2000 processQueue: exceptions.IOError / [Errno 24] Too many open files: '/var/lib/mailman/data/mm_q.29101.1'
It seems that for each log entry run_queue opens the logfile, and then doesn't close it. Is there a quick fix for this somewhere ? I'm running mailman 1.1
-- Madarasz Gergely gorgo@caesar.elte.hu gorgo@linux.rulez.org It's practically impossible to look at a penguin and feel angry. Egy pingvinre gyakorlatilag lehetetlen haragosan nezni. HuLUG: http://mlf.linux.rulez.org/

"GM" == Gergely Madarasz <gorgo@caesar.elte.hu> writes:
GM> It seems that for each log entry run_queue opens the logfile,
GM> and then doesn't close it. Is there a quick fix for this
GM> somewhere ? I'm running mailman 1.1
I seem to remember someone posting a patch for this, so check the archives. Otherwise, don't sweat it, run_queue is gone, gone, gone in the CVS version.
-Barry

On Thu, 10 Feb 2000, Barry A. Warsaw wrote:
"GM" == Gergely Madarasz <gorgo@caesar.elte.hu> writes:
GM> It seems that for each log entry run_queue opens the logfile, GM> and then doesn't close it. Is there a quick fix for this GM> somewhere ? I'm running mailman 1.1
I seem to remember someone posting a patch for this, so check the archives. Otherwise, don't sweat it, run_queue is gone, gone, gone in the CVS version.
Hmm... searching the archives I found the following which might be of help:
[Mailman-checkins] CVS: mailman/Mailman LockFile.py Update of /projects/cvsroot/mailman/Mailman In directory anthem:/home/bwarsaw/projects/mailman/Mailman Modified Files: LockFile.py Log Message: Use one module global ... http://www.python.org/pipermail/mailman-checkins/1999-December/ 002573.html - size 2.0K 47% 14 Dec 99
but the mailman chekins directory doesn't seem to be there...
-- Madarasz Gergely gorgo@caesar.elte.hu gorgo@linux.rulez.org It's practically impossible to look at a penguin and feel angry. Egy pingvinre gyakorlatilag lehetetlen haragosan nezni. HuLUG: http://mlf.linux.rulez.org/
participants (2)
-
Barry A. Warsaw
-
Gergely Madarasz