From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?iso-8859-1?Q?Jakob_=D8stergaard?= Subject: Re: raid 5 question Date: Tue, 30 Apr 2002 22:19:36 +0200 Sender: linux-raid-owner@vger.kernel.org Message-ID: <20020430221936.E15847@unthought.net> References: <20020429041335.A12337@unthought.net> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: ; from derek@cynicism.com on Mon, Apr 29, 2002 at 10:05:12AM -0700 To: Derek Vadala Cc: linux-raid@vger.kernel.org List-Id: linux-raid.ids On Mon, Apr 29, 2002 at 10:05:12AM -0700, Derek Vadala wrote: > On Mon, 29 Apr 2002, [iso-8859-1] Jakob =D8stergaard wrote: >=20 > > raidreconf (http://unthought.net/raidreconf) will do what you are a= sking, >=20 > Is that the official site and are you and/or Danny Cox still working = on > the code?=20 1) It is the official site 2) We're not working on the code ;) >I noticed that the raidreconf bundled with the raidtools-1.0 in > skipjack is 0.1.2 and the one at unthought.net is 0.1.1-- but maybe t= hat's > because someone else worked on it and didn't put their name in. That is interesting - too bad RedHat didn't care to ask me about the kn= own problems with the tool... Some of the known problems are easily fixable= =2E I guess their customers will tell them what happens when raidreconf enc= ounters a bad block ;) (trivial fix by the way) >=20 > > In any case, I would be interested in hearing your experiences if y= ou > > decide to try out raidreconf. >=20 > I've still got a bunch of comparison tests to run, but I'll post the > results here when I' m done. Great ! Thanks, Oh, and just to let everyone know: I'm hoping that someone is going to= take raidreconf and integrate it into LVM or EVMS - this is clearly the way = that things *should* be going, and it would be needed for hot-reconfiguratio= n anyway. I can't do this myself, because I simply don't have the time. As I see it - while raidreconf might be useful as it is today (I wrote = it because I needed it, and it has helped me a few times), the concept of = having to dismount, raidstop, reconfigure (for 20+ hours!), raidstart, ext2res= ize, and remount, is dead. It is simply not acceptable. RAID reconfiguration *= MUST* be on-line, and it must be something that's just running in the background= while the server is ticking away doing whatever it is it is doing. Knowing that I can't integrate it in the kernel as it should be (becaus= e of my own time constraints), and knowing that the way the tool works today (userspace, non on-line) is not acceptable, doesn't give me much of an incentive to keep "working" on it. Now I will try to push this to someone else, rather than feeling bad ab= out not doing anything about a potentially really useful tool myself. Cheers, :) --=20 =2E............................................................... : jakob@unthought.net : And I see the elder races, : :.........................: putrid forms of man : : Jakob =D8stergaard : See him rise and claim the earth, : : OZ9ABN : his downfall is at hand. : :.........................:............{Konkhra}...............: - To unsubscribe from this list: send the line "unsubscribe linux-raid" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html