From owner-freebsd-hackers Sat Sep 16 18:08:34 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id SAA11188 for hackers-outgoing; Sat, 16 Sep 1995 18:08:34 -0700 Received: from ns1.win.net (ns1.win.net [204.215.209.3]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id SAA11179 for ; Sat, 16 Sep 1995 18:08:31 -0700 Received: (from bugs@localhost) by ns1.win.net (8.6.11/8.6.9) id VAA02600 for hackers@freebsd.org; Sat, 16 Sep 1995 21:14:51 -0400 From: Mark Hittinger Message-Id: <199509170114.VAA02600@ns1.win.net> Subject: Re: FreeBSD Max Users (fwd) To: hackers@freebsd.org Date: Sat, 16 Sep 1995 21:14:50 -0400 (EDT) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 1025 Sender: owner-hackers@freebsd.org Precedence: bulk > From: Andreas Klemm > The initial value of 10 isn't that bad. 250 is _overkill_. It makes > certain kernel structures too huge and _kills_ your performcance. > I'd choose values of 10-20 depending on the amount of memory > you have and the number of users, that will work at the same time > on your system. I don't want to offend Andreas but I think this information might not be entirely accurate. I've been using a setting of 256 for around 8 months now. Creating a large swap area, and using a larger maxusers parameter has been necessary for me to operate my production internet servers. I do not see a performance hit from this. I do see a major performance hit when I don't use enough swap space or if I don't pump up certain kernel parameters - the system either hangs or crashes! Don't be shy, pump it up and see what happens. Experiment. It is ok to do this with FreeBSD - none of us will mind :-) Regards, Mark Hittinger Internet Manager WinNET Communications, Inc. bugs@win.net