Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 3 May 2001 11:04:59 +0930
From:      Greg Lehey <grog@lemis.com>
To:        Douglas Egan <degan@calcon.net>
Cc:        Brian Skrab <bgs@pinky.us.net>, freebsd-questions@FreeBSD.ORG
Subject:   Re: Vinum & FreeBDS 4.3-stable???
Message-ID:  <20010503110459.O72846@wantadilla.lemis.com>
In-Reply-To: <3AF0B492.8AB9DCCB@calcon.net>; from degan@calcon.net on Wed, May 02, 2001 at 08:29:55PM -0500
References:  <Pine.BSF.4.32.0105021459520.326-100000@pinky.us.net> <20010503084922.B72846@wantadilla.lemis.com> <3AF0B492.8AB9DCCB@calcon.net>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wednesday,  2 May 2001 at 20:29:55 -0500, Douglas Egan wrote:
> Does anyone have any comparisons or comments on RAID5 implemented
> with vinum and FreeBSD vs RAID5 implemented with RAIDFrame and
> NetBSD?

No direct comparisons that I know of.  I read a thread on the NetBSD
mailing lists a day or two ago which suggested that it's possible to
(mis)configure RAIDFrame to perform very, very badly.  I don't know of
a way to do this with Vinum.  In general, Vinum has a reputation for
very good performance.

> My nfs/smb server is currently running NetBSD 1.5 with RAID5 using
> RAIDFrame.  When I configured that system, I was unaware of vinum's
> support of RAID5.
>
> Can you comment on vinums stability with RAID5?  I have currently
> configured a 4 disk array; 3 15gig ATA ultra66 and 1 18gig scsi.
> This is split up into 2 22gig arrays with no hot-spares.  So far the
> RAIDFrame has performed admirably.

We have had some problems with RAID-5 stability in the past.  The only
current issue is with memory mapped data: if the data changes while
it's being written, there's a potential for data corruption.  This is
a theoretical issue which I'm currently considering: I don't know if
it's possible for the data to change while it's being written.

You wrote this in reply to a message from Brian Skrab reporting
extreme data corruption.  I don't know what the problem here is, but
it's the first report of anything like this that I've seen.

Greg
--
When replying to this message, please copy the original recipients.
If you don't, I may ignore the reply.
For more information, see http://www.lemis.com/questions.html
Finger grog@lemis.com for PGP public key
See complete headers for address and phone numbers

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?20010503110459.O72846>