Date: Sat, 28 Feb 1998 13:42:28 +1030 From: Greg Lehey <grog@lemis.com> To: shimon@simon-shapiro.org Cc: hackers@FreeBSD.ORG, blkirk@float.eli.net, jdn@acp.qiv.com, Wilko Bulte <wilko@yedi.iaf.nl> Subject: Re: SCSI Bus redundancy... Message-ID: <19980228134228.40320@freebie.lemis.com> In-Reply-To: <XFMail.980227170450.shimon@simon-shapiro.org>; from Simon Shapiro on Fri, Feb 27, 1998 at 05:04:50PM -0800 References: <19980228110827.36052@freebie.lemis.com> <XFMail.980227170450.shimon@simon-shapiro.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 27 February 1998 at 17:04:50 -0800, Simon Shapiro wrote: > > On 28-Feb-98 Greg Lehey wrote: >> On Fri, 27 February 1998 at 15:20:55 -0800, Simon Shapiro wrote: >>> >>> On 26-Feb-98 Wilko Bulte wrote: >>>> As Simon Shapiro wrote... >>>>> >>>>> On 25-Feb-98 Wilko Bulte wrote: >>>>> ... >>>>> >>>>>> Digital Unix TruClusters do DRD (distributed raw device) now. Things >>>>>> like Oracle Parallel Server love this. A cluster filesystem is >>>>>> another >>>>>> kettle of fish of course. But not impossible, see OpenVMS. >>>>> >>>>> Stay tuned... FreeBSD will have this functionality too. >>>> >>>> Next step: a volume manager? >>> >>> I'll let someone else commit to this one. >> >> Done. >> >>> 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? > > Which? > > Julian's stuff? I do not clearly remember, but a slice can be re-written > and then re-evaluated. Care needs to be excercised in not overlapping, not > destroying things, etc. No, sorry, I wasn't talking about Julian's stuff (which I *still* haven't looked at; sorry, Julian). > DPT arrays? Simple; you make an ioctl call into the DPT driver, I > write a message to the controler, specifying which disk to add to > which array, the controller starts a hot rebuild, etc. The details > escape me right now, but I belive it is doable. Why would you want > to do that? No idea... :-) I was thinking more about what the ioctl call does. Raid 5 spreads data in stripes over all drives (or subdisks if you're using a Veritas-like volume manager). Add a drive, and you break this mapping. The only way to rearrange is to rewrite all the disks. Is that what they do? Or are they cheating in some manner? Greg 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?19980228134228.40320>