Date: Sat, 28 Feb 1998 02:35:36 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: grog@lemis.com (Greg Lehey) Cc: shimon@simon-shapiro.org, wilko@yedi.iaf.nl, jdn@acp.qiv.com, blkirk@float.eli.net, hackers@FreeBSD.ORG Subject: Re: SCSI Bus redundancy... Message-ID: <199802280235.TAA22897@usr06.primenet.com> In-Reply-To: <19980228110827.36052@freebie.lemis.com> from "Greg Lehey" at Feb 28, 98 11:08:27 am
next in thread | previous in thread | raw e-mail | index | archive | help
> > I think Julian's SLICE code has something in that direction. DPT
> > supports INCREASING the size of a RAID-5 array by adding drives.
>
> How can that work?
Take a set of devices:
[ ]
[ ]
[ ]
The kernel views these as linear arrays of blocks.
Steal the first block and/or cylinder and put:
struct ccd_label {
char signature[ 20]; /* "This is a CCD device"*/
time_t timestamp; /* set created this time*/
long ident; /* set's additional uniquifier*/
long item; /* piece number ...*/
long of; /* of...*/
};
on the front of it.
Then you have a CCD SLICE manager that claims these things, and when
all N item's of an N item device "arrive", it exports another device
that's the agregate of N devices.
It works because the SLICE code implements "arrival" events.
Personally, I want to use "arrival" events to trigger mounts. 8-).
The mapping into the FS hierarchy is a seperate problem altogether.
Combine this with "soft readonly" to go with you "soft updates",
and you should be able to just turn things off and turn them on
and not need fsck's or have mount problems, etc..
Terry Lambert
terry@lambert.org
---
Any opinions in this posting are my own and not those of my present
or previous employers.
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?199802280235.TAA22897>