Date: Tue, 30 Mar 1999 14:13:45 -0500 From: Robert Sexton <robert@kudra.com> To: Wilko Bulte <wilko@yedi.iaf.nl> Cc: freebsd-hackers@freebsd.org Subject: Re: another ufs panic.. Message-ID: <19990330141345.A22153@kudra.com> In-Reply-To: <199903301840.UAA01034@yedi.iaf.nl>; from Wilko Bulte on Tue, Mar 30, 1999 at 08:40:20PM %2B0200 References: <3.0.6.32.19990330083127.007bf2f0@192.168.255.1> <199903301840.UAA01034@yedi.iaf.nl>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Mar 30, 1999 at 08:40:20PM +0200, Wilko Bulte wrote: > As Bob Bishop wrote ... > > At 17:53 29/03/99 -0800, Matthew Dillon wrote: > > > Generally speaking, glitches will cause SCSI bus parity errors. Maybe > > "Generally" that is true... ;-) I'm just playing the devil's advocate > a bit here. > > > > not all glitches, but a high enough percentage that you usually get > > > some sort of indication that there is a problem. > > True. But I've heard stories about PCI bridge chips in combination > with specific adapter firmware revs corrupting data. In general, everything in > the datapath can cause you grief. With the crappy PC hardware out there > (and/or overclocked crappy PC h/w to make it worse) anything is possible. > Even without SCSI. I saw an actual example of this last month. SymBIOS controller, FIC motherboard. The supplier had flashed the wrong bios onto the machine, and hard drive data was quickly scrambled without any SCSI errors at all. Linux died quickly. FreeBSD lasted longer, but eventually threw in the towel due to all of the data corruption. -- Robert Sexton, robert@kudra.com "No one told me that it could not be done, and so I did it." - Jack Kloepfer Read the Newton FAQ! <http://www.kudra.com/newton/newton-faq>; To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19990330141345.A22153>