From owner-freebsd-hackers Wed Jan 27 01:26:23 1999 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA19547 for freebsd-hackers-outgoing; Wed, 27 Jan 1999 01:26:23 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from iquest3.iquest.net (iquest3.iquest.net [209.43.20.203]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id BAA19542 for ; Wed, 27 Jan 1999 01:26:21 -0800 (PST) (envelope-from toor@y.dyson.net) Received: (qmail 4360 invoked from network); 27 Jan 1999 09:26:18 -0000 Received: from dyson.iquest.net (HELO y.dyson.net) (198.70.144.127) by iquest3.iquest.net with SMTP; 27 Jan 1999 09:26:18 -0000 Received: (from root@localhost) by y.dyson.net (8.9.1/8.9.1) id EAA00689; Wed, 27 Jan 1999 04:26:16 -0500 (EST) Message-Id: <199901270926.EAA00689@y.dyson.net> Subject: Re: High Load cron patches - comments? In-Reply-To: <199901262256.OAA22114@apollo.backplane.com> from Matthew Dillon at "Jan 26, 99 02:56:10 pm" To: dillon@apollo.backplane.com (Matthew Dillon) Date: Wed, 27 Jan 1999 04:26:16 -0500 (EST) Cc: wes@softweyr.com, toasty@home.dragondata.com, hackers@FreeBSD.ORG From: "John S. Dyson" Reply-To: dyson@iquest.net X-Mailer: ELM [version 2.4ME+ PL38 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Matthew Dillon said: > > Sendmail's load average-baesd regulation doesn't work worth beans, > but putting a max limit on the number of forked children and queue-running > children solved nearly all of sendmail's load-related problems. > Get your old DSP textbooks out: Create a new filter (instead of the very slow attack/decay load average), with a fast attack (but maybe not a process fork limit, because it is too static), and fairly slow decay. If you need help, I can write a filter, and the current 10Hz clock is probably quick enough. The algorithm would be similar to my new VM limit code (which I haven't released to anyone yet.) There is a weighted peak limit at the sample rate, but it is averaged in such a way that peak limit doesn't mess things up like a hard process (or VM page count) limit would. -- John | Never try to teach a pig to sing, dyson@iquest.net | it makes one look stupid jdyson@nc.com | and it irritates the pig. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message