Date: Thu, 27 Sep 2001 01:29:55 -0600 From: Samuel Greear <dragonk@evilcode.net> To: Greg Lehey <grog@FreeBSD.org>, Jason Zawacki <jzawacki@appliedtheory.com> Cc: freebsd-questions@FreeBSD.org Subject: Re: vinum IDE RAID5 troubles Message-ID: <01092701295503.19839@beware.dragonknight.net>
next in thread | raw e-mail | index | archive | help
On Wednesday 26 September 2001 06:45 pm, Greg Lehey wrote: > > The 7 drives are connected to various headers on the Promise cards > > (ie, none are connected to the HPT Headers on the motherboard, since > > I've had about a billion problems with it, even with the latest HPT > > BIOS (1.28)). > > I have a nearly identical machine, BP6, dual celery 400, 512MB, etc. I have a 45 gig WD drive on each HPT channel in a vinum stripe that has been performing flawlessly under fairly heavy load for over 6 months. > > Now, the problems I've been happen after successfully getting the > > RAID5 volume up and running (setting up the volume, initting it, and > > newfsing it). Under heavy I/O (when xferring a 4GB file over at > > about 2MB/s) it gets about halfway through the xfer, and then one of > > the controllers will time out and jump down to PIO mode, and then > > the rest follow. After this happens, at a random time after, the > > box freezes up resulting in huge data corruption on the volume. > > Also, one of the drives always goes stale, requiring me to start it > > again which sometimes results in a freeze after a while. > Did you try rebooting the box right after setting up the volume before you start copying things over and testing it? I've had problems when not rebooting first, panic's, etc. But a simple reboot before using the volume and everything works peachily so I haven't attempted to debug... -- Samuel J. Greear <dragonk@evilcode.net> Developer - GetMegabits, Inc. http://www.itmom.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?01092701295503.19839>