From owner-freebsd-hackers Thu Apr 6 09:27:46 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id JAA01463 for hackers-outgoing; Thu, 6 Apr 1995 09:27:46 -0700 Received: from cs.weber.edu (cs.weber.edu [137.190.16.16]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id JAA01457 for ; Thu, 6 Apr 1995 09:27:45 -0700 Received: by cs.weber.edu (4.1/SMI-4.1.1) id AA12169; Thu, 6 Apr 95 10:15:51 MDT From: terry@cs.weber.edu (Terry Lambert) Message-Id: <9504061615.AA12169@cs.weber.edu> Subject: Re: large filesystems/multiple disks [RAID] To: babkin@hq.icb.chel.su (Serge A. Babkin) Date: Thu, 6 Apr 95 10:15:50 MDT Cc: freebsd-hackers@FreeBSD.org In-Reply-To: <199504041628.LAA00977@hq.icb.chel.su> from "Serge A. Babkin" at Apr 4, 95 11:28:40 am X-Mailer: ELM [version 2.4dev PL52] Sender: hackers-owner@FreeBSD.org Precedence: bulk > > But, again, the risk is increased by the spanning, which was the point > > of the post... that spanning is nearly useless without additional > > support changes to increase reliability. > > How about disk mirroring ? If you have a _mirrored_ database on two 2G > disks and then add 2x2G (or may be 1x4G) and get spanned mirrored database. > You get reliability due to mirroring and easy expansion due to spanning. That works... but then again, it's an additional support change to increase reliability. The minimum number of changes for spanning is two: the first to get spanning, the second to increase reliability so you can actually use the spanning safely. Note that a procedural change could suffice... like adding regular backups when there were none before. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.