From owner-freebsd-stable@FreeBSD.ORG Wed Jun 11 03:04:02 2003 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 8F80537B401 for ; Wed, 11 Jun 2003 03:04:02 -0700 (PDT) Received: from anduin.net (anduin.net [212.12.46.226]) by mx1.FreeBSD.org (Postfix) with SMTP id 7C57943FAF for ; Wed, 11 Jun 2003 03:04:01 -0700 (PDT) (envelope-from ltning@anduin.net) Received: (qmail 71673 invoked from network); 11 Jun 2003 10:03:58 -0000 Received: from celduin.net (HELO ?192.168.1.10?) (81.0.162.106) by anduin.net with SMTP; 11 Jun 2003 10:03:58 -0000 From: Eirik Oeverby To: stable@freebsd.org In-Reply-To: <3EE6FD38.C8AC3358@dolaninformation.com> References: <6906683B-9AB7-11D7-A8ED-000A956789F6@katamail.com> <3EE5D24A.BFE58FA6@dolaninformation.com> <1055252312.49655.26.camel@ranger.anduin.net> <3EE6FD38.C8AC3358@dolaninformation.com> Content-Type: text/plain Message-Id: <1055325835.266.6.camel@ranger.anduin.net> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.3.92 (Preview Release) Date: 11 Jun 2003 12:03:55 +0200 Content-Transfer-Encoding: 7bit Subject: Re: A stupid question... X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 11 Jun 2003 10:04:02 -0000 Hi, I know about the buffer issues - I've had that on all my servers, and nowadays it's routine to up that to atleast 4x the default. So that's not the problem. There are no messages in any logs, apart from dropped connections due to the outage. I'm fairly sure it's the NIC itself, or the combination of NIC and motherboard. It's a non-important server and the NIC is only a backup NIC anyway (which I sometimes abuse to speed up LAN traffic) - so I won't be investing any more time on that. It's about to be decommissioned soon. Thanks for the ideas, though. /Eirik On Wed, 2003-06-11 at 11:58, Greg Panula wrote: > Eirik Oeverby wrote: > > > > Hi, > > > > I have the rl problem from time to time - it is as if the interface is > > unplugged from the HUB. However, I have no problem pinging it from the > > host itself, but it's gone to anyone on the outside. > > A "ifconfig rl0 DOWN ; ifconfig rl0 UP" usually solves the problem until > > it happens next time - which can be minutes, hours, days or weeks later. > > I suppose in average I have it happening two to three times per month. > > On another machine with the same FreeBSD version I do not have the same > > problem, but the hardware is entirely different. Only the NIC driver > > used is the same. > > > > Check network buffers 'netstat -m'. If peak equals max, you should bump > up the amount of memory set aside for network buffers. See 'man 7 > tuning' for information about increasing network buffers. > > Check your logs(e.g. /var/log/messages) for any interesting > messages&hints... maybe something related to rl0 losing connectivity, > buffer space, etc. > > You might also want to try swapping out the nic. RealTek nics are > fairly inexpensive. > > good luck, > greg