From owner-freebsd-current@FreeBSD.ORG Mon Nov 12 21:43:48 2007 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 7EF5916A421 for ; Mon, 12 Nov 2007 21:43:48 +0000 (UTC) (envelope-from askbill@conducive.net) Received: from conducive.net (lindfield.ch [203.194.153.81]) by mx1.freebsd.org (Postfix) with ESMTP id 43EF813C4B0 for ; Mon, 12 Nov 2007 21:43:48 +0000 (UTC) (envelope-from askbill@conducive.net) Received: from cm218-253-81-177.hkcable.com.hk ([218.253.81.177]:64273 helo=pb.local) by conducive.net with esmtpsa (TLSv1:AES256-SHA:256) (Exim 4.63 (FreeBSD)) (envelope-from ) id 1Irh43-000O5P-4o for freebsd-current@freebsd.org; Mon, 12 Nov 2007 21:43:27 +0000 Message-ID: <4738C8FE.7030402@conducive.net> Date: Mon, 12 Nov 2007 21:43:26 +0000 From: =?UTF-8?B?6Z+T5a625qiZIEJpbGwgSGFja2Vy?= User-Agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.8.1.2) Gecko/20070221 SeaMonkey/1.1.1 MIME-Version: 1.0 To: freebsd-current@freebsd.org References: <47375D78.4040309@chuckr.org> <20071112005620.GA87383@cdnetworks.co.kr> <4737BCC3.5030509@chuckr.org> <20071112044340.GC87383@cdnetworks.co.kr> <4738B818.2070204@chuckr.org> <4738BDA4.3070405@conducive.net> <4738C36B.9090302@chuckr.org> In-Reply-To: <4738C36B.9090302@chuckr.org> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Subject: Re: nfe internet problems X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 12 Nov 2007 21:43:48 -0000 Chuck Robey wrote: > 韓家標 Bill Hacker wrote: >> Chuck Robey wrote: >> >> *snip* >>> >>> I have one other odd problem, it's that no probe messages whatever >>> print during booting, so I had to pick up the demsg later on. The >>> last thing I see during booting is the first spinner character, >>> usually "|", and the next thing I see is "Login;", nothing inbetween >>> whatever. Any idea why my booting has gone totally silent on me? >> >> Not so odd. This is the way we set up servers. '-h' for headless. >> >> Now - the question is *why* IF/AS/WHEN you have no such setting. >> >> My guess is BIOS reporting one thing, probe another w/r finding a VGA >> it is allowed to use for console. >> >> What the probe 'discovers' doen't go active until it has finished - >> but the first two loader stages simply 'assume' console will be dealt >> with somehow, so their output does show up - boot device choice, >> spinner and such are still BIOS turf. >> > > Hmm I haven't given as muich care to my linker.hints file as I could. I > think I will experiment with giving it, at least, the correct stuff for > the console, that might give me the probe messages back. > > Good hint, I think this one's likely to fix the probing/printing > problem, but what about the problem with my nfe ports? If I activate > them with ifconfig, it locks up my computer, that's kind of a mean way > to react ... > Try the plastic soda straw trick. Search the kitchen for a medium soda straw. Liberate a Berg jumper, preferably the sort with a pull-tab tail. Insert into one end of the soda straw, staple in place, wrap a turn of Scotch tape around it. Now - unlike that old set of Kelly forceps from S-100 days, you can clear the MB CMOS w/o contortions, removing cards, or short-circuiting add-on cards. Why so? Modern MB with soft power-off keep an amp or more of power running about all over the board and bus in the power-off state. This for power-on by wake-on-LAN (card has to be 'alive', no?), wake-on-keypress, and a host of other stuff. Removing a card, attaching a cable, powering up with a not-fully-seated (either) *now and then* creates a spike that smacks a bit - or several - sidegodlin inside the BIOS alleged-NV RAM. Too often these are interpreted as 'settings' that the BIOS UI could not set by menu, nor do they make good sense to the near-as-dammit 'state machine' the BIOS is supposed to present for later configuration as a proper Von Neumann architecture Turing machine. But it tries.... So... Pull the mains cord and do a five or ten count before attaching/detaching. inserting/removing. Use yer soda-straw *in the mains-disconnected state* to do a 'hard' CMOS clear and start fresh when wonkiness of the sort you are seeing appears. Don't be surprised that the issue goes away. Unless you have zapped the glue chips. Which also happens 'now and then'. 'Specially on cold, dry, days with static build-up. Chipsets installed in a MB have pretty good protection. But a lightening bolt - even nano-sized, is a formidable motivator. HTH, Bill