From owner-freebsd-current@FreeBSD.ORG Wed Jan 7 01:21:38 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 65C9616A4CE; Wed, 7 Jan 2004 01:21:38 -0800 (PST) Received: from gate.bitblocks.com (bitblocks.com [209.204.185.216]) by mx1.FreeBSD.org (Postfix) with ESMTP id CAC8043D58; Wed, 7 Jan 2004 01:21:34 -0800 (PST) (envelope-from bakul@bitblocks.com) Received: from bitblocks.com (localhost [127.0.0.1]) by gate.bitblocks.com (8.12.10/8.12.10) with ESMTP id i079LVic070520; Wed, 7 Jan 2004 01:21:32 -0800 (PST) (envelope-from bakul@bitblocks.com) Message-Id: <200401070921.i079LVic070520@gate.bitblocks.com> To: "Daniel O'Connor" In-reply-to: Your message of "Wed, 07 Jan 2004 17:31:40 +1030." <200401071731.40481.doconnor@gsoft.com.au> Date: Wed, 07 Jan 2004 01:21:31 -0800 From: Bakul Shah cc: Don Lewis cc: freebsd-current@freebsd.org cc: yurtesen@ispro.net.tr Subject: Re: FreeBSD Crashes with AMD X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 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: Wed, 07 Jan 2004 09:21:38 -0000 > > for example, > > http://www.anime.net/~goemon/linux-ecc/ > > Though, not all are built well enough. > > Hmm, well I am very suprised to see the Via VT400 in there - I can't see > anything on the spec page for it that says it supports ECC, and the manual > for the KT400 board doesn't say explicitly that ECC is supported. I haven't checked the veracity of information on that page. Also see tomsHardware site. My point was ECC systems are available and not too expensive anymore. But you'd have to check a vendor's claim before buying. > It would be nice, but there are patches out there, grab them, clean them up > and submit them :) If anyone else gets there before me I wouldn't be upset! But seriously, while I understand someone has to sign up for it, a server class OS needs to have ecc support. You can hand out such jobs the next time someone new asks how they can help. But you'd have to clearly specify the task (a man page is usually sufficient). > I just thought my idea was pretty cute, it would also be nice to say to people > with mystery SIGSEGV's that the break into the loader type 'memtest' and see > if they get errors :) It was creative alright! My experience has been that memtest like tests do not help with nasty, marginal power/timing related errors that only up on a heavily loaded multiuser os. Start a few compiles, finds, pure number crunching programs, throw in a few crashme kind of tests and see how well things stand up. Then run the same load at highest/lowest rated temperatures and for 24 hours or more.