From owner-freebsd-net Thu Mar 25 14:32: 5 1999 Delivered-To: freebsd-net@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [209.157.86.2]) by hub.freebsd.org (Postfix) with ESMTP id D63EE14EA8; Thu, 25 Mar 1999 14:31:57 -0800 (PST) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id OAA03635; Thu, 25 Mar 1999 14:31:28 -0800 (PST) (envelope-from dillon) Date: Thu, 25 Mar 1999 14:31:28 -0800 (PST) From: Matthew Dillon Message-Id: <199903252231.OAA03635@apollo.backplane.com> To: Jesse Cc: Luigi Rizzo , freebsd-net@FreeBSD.ORG, freebsd-hackers@FreeBSD.ORG Subject: Re: 3.1-STABLE dies on 40+ connects References: Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org :Just to provide a little bit more info, Alex, one of the developers of :icecast said that a person running OpenBSD had kernel panics under similar :conditions but that it was never resolved. : :The machine is remote, but I can have someone stand by and record the :kernel panic message while I crash it, if you it'd be useful (probably). : :--- :Jesse :http://www.lumiere.net/ On the face of it it sounds to me like icecast is perhaps making the socket buffers bigger --- maybe too big. The default socket buffer size is 16K receiving, 16K sending. 40 x 16K x 2 = 1MB. If icecast is increasing the size of the socket buffers, this value can explode. netstat -tn on a running system ought to give you some idea about that by seeing how backed-up the socket buffers get. -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message