From owner-freebsd-stable Tue Feb 22 21:27:27 2000 Delivered-To: freebsd-stable@freebsd.org Received: from phoenix.welearn.com.au (phoenix.welearn.com.au [139.130.44.81]) by hub.freebsd.org (Postfix) with ESMTP id 68C0537B93D for ; Tue, 22 Feb 2000 21:27:14 -0800 (PST) (envelope-from jon@phoenix.welearn.com.au) Received: (from jon@localhost) by phoenix.welearn.com.au (8.9.3/8.9.3) id QAA67928 for freebsd-stable@FreeBSD.ORG; Wed, 23 Feb 2000 16:26:57 +1100 (EST) (envelope-from jon) Date: Wed, 23 Feb 2000 16:26:54 +1100 From: Jonathan Michaels To: "'freebsd-stable'" Subject: Re: Signal 11 error when installing 3.4 STABLE Message-ID: <20000223162653.A67290@phoenix.welearn.com.au> Reply-To: jon@welearn.com.au Mail-Followup-To: 'freebsd-stable' References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.95.4i In-Reply-To: ; from Alok Dhir on Tue, Feb 22, 2000 at 11:29:56AM -0500 Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Tue, Feb 22, 2000 at 11:29:56AM -0500, Alok Dhir wrote: > If I could add one thing - its typically difficult to accept that these > problems are caused by hardware. "But my machine runs Win9x perfectly" is a > typical retort when the finger is pointed at hardware. > > It is important to note that FreeBSD/Linux/etc are significantly more > demanding than Win9x, etc, and are much more likely to bring flaky hardware > problems to the surface. > > I've seen Signal 11 many times over the past several years from various > versions of FreeBSD and LInux, and so far, it has been caused by bad > hardware (usually memory) every time... another point of view .. or, the last (first and only) time i had serous sig11's was when i ran freebsd+x11+several linux apps in the linux emulator. my problems started when i moved from freebsd-release v2.1.5 to freebsd v2.1.7-release, it was a reproducable fault and only happend with freebsd+x11+linux apps running in the linux emulator. at the time, i had had tested the hardware with ibm os/2 (under load) and a basic qnx installation (with several load tests involving qnx windows -- like x11 but quicker cleaner more reliable etcetc) ... and having the memory replaced 3 times and the rest of teh hardware tested (it is supermicro motherboard, p6sne). i nearly was thrown out by the sech support team who couldn't find any out of spec issues with my hardware. i;d been dealing this that crew for some 10 years and they knew how pedantic i could be about 'out of spec' hardware, but over this issue i nearly destroyed that relationship .. well so many experienced freebsd users can't all be wrong ... that time they were, all of them. cutting a long and made needlessly bitter buy the less than informed comment on the mailinglist the probelm turned out to be the linux code running on in linux emulator that convinced the freebsd kernel that it was a segmentation fault -- sig11. what needs to be stressed in these gotta find a quick fix situations is that the fault is not always the forst most obvious choice in the multiple choice presentation list. i'm not a programmer, i'm no longer as good with hardware as i used to be .. but it was clear to me that the voluminious chorus that kept ranting memory failure really should have checked what they were saying .. it still hurts even 2 years down the track. we all need to take care of the how and the what we say not all people brag about what they have done, where they have been or what qualifications they may or may not have .. its not always possible to see beyound the words on the screen, especially if the person has some handicap, be it esl (english as a second language) some toerh midically regognised 'disability' or just a but hunngover from teh night before. we all need to be aware of what we re saying, especially when it comes to the airy fairy faults that most sig11 seem to be, we all need to listen to what is being said, not to what we would like to have been said, regarding my detailed explanation of the process i'd gone through to prove that it could not have been a bsaic hardware fault and that it was in fact a software and in particular a freebsd fault. i realise that most operating system users are failry one eye'd but that one eyedness should'nt blind us to the fact that freebsd is not the only fish in teh bowl, neither is it the be all and end all the way so called reliable and or bullet proof systems are built. sorry i'll get of my soap box now. i'm just asking that we call to the voice of reason, and an ear of compassion when listening to someone struggling to expalin a probelm that is hard for the particular to explain things, concepts that my be as foreign to that person as say bing disabled is to most of the participant in these mailing lists. regards jonathan -- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message