From owner-freebsd-net Mon Feb 12 6:46:25 2001 Delivered-To: freebsd-net@freebsd.org Received: from prism.flugsvamp.com (cb58709-a.mdsn1.wi.home.com [24.17.241.9]) by hub.freebsd.org (Postfix) with ESMTP id 6895137B401; Mon, 12 Feb 2001 06:46:22 -0800 (PST) Received: (from jlemon@localhost) by prism.flugsvamp.com (8.11.0/8.11.0) id f1CEkvH89288; Mon, 12 Feb 2001 08:46:57 -0600 (CST) (envelope-from jlemon) Date: Mon, 12 Feb 2001 08:46:57 -0600 From: Jonathan Lemon To: Alfred Perlstein Cc: net@freebsd.org, jlemon@freebsd.org Subject: Re: somaxconn and foot removal Message-ID: <20010212084657.U650@prism.flugsvamp.com> References: <20010211015516.J3274@fw.wintelcom.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0pre2i In-Reply-To: <20010211015516.J3274@fw.wintelcom.net> Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sun, Feb 11, 2001 at 01:55:16AM -0800, Alfred Perlstein wrote: > The sysctl for somaxconn is an int, however the queue limits in the > socket structures are 'short' this can cause some bad behavior if > one sets somaxconn to more than 32k. > > A) So, do we bump the sockets to use 'int' for so->so_qlimit? > B) Do we fix solisten() to compensate? > C) Or de we fix the sysctl (patch below)? I'd patch sysctl. I can't imagine a listen queue > u_short actually being all that useful; at some point it is better to just start shedding load instead of queuing up numerous connections. -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message