From owner-freebsd-stable Fri Mar 3 5:30: 7 2000 Delivered-To: freebsd-stable@freebsd.org Received: from peak.mountin.net (peak.mountin.net [207.227.119.2]) by hub.freebsd.org (Postfix) with ESMTP id D915D37B566 for ; Fri, 3 Mar 2000 05:30:01 -0800 (PST) (envelope-from jeff-ml@mountin.net) Received: (from daemon@localhost) by peak.mountin.net (8.9.1/8.9.1) id HAA14758; Fri, 3 Mar 2000 07:29:49 -0600 (CST) (envelope-from jeff-ml@mountin.net) Received: from dial-90.max1.wa.cyberlynk.net(207.227.118.90) by peak.mountin.net via smap (V1.3) id sma014755; Fri Mar 3 07:29:27 2000 Message-Id: <3.0.3.32.20000303072732.00989b20@207.227.119.2> X-Sender: jeff-ml@207.227.119.2 X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.3 (32) Date: Fri, 03 Mar 2000 07:27:32 -0600 To: Jamie Gritton , Khetan Gajjar From: "Jeffrey J. Mountin" Subject: Re: recent vinum changes Cc: stable@FreeBSD.ORG In-Reply-To: References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG At 01:19 PM 3/2/00 -0700, Jamie Gritton wrote: >Khetan Gajjar writes: > >> How are you reviving a vinum mirror ? I haven't yet had the >> need to do so, but I'm dreading doing it now that you mention >> this. > > Generally, a "vinum start" on the previously dead plex. It will usually >revive OK if you do this in single-user, but any reasonable level of >filesystem activity panics the system, and when it comes up, there's nothing >left of /usr (on either half of the mirror). What was the previous state of the dead plex and it's subdisks? I've had subdisks go "stale" or even "faulty" by themsleves or by design (ala testing), but the plex is only "down" (if that's what you mean by "dead"). Only problems I've had when reviving were due to pilot error, so must have stepped right on the bug that was fixed in the recent commits. >> What happens if one of the disks dies ? Does the machine >> panic ? > > No, it does OK as long as you never try to revive. That's our current >situation. From your first message the loss of data on the "up" plex when reviving isn't reassuring, but I've blown up plexes various ways and recovered most every time. Note the "most" covers the more exotic and crazy ideas. I'd be some of the commits will filter back in a week or 6 (the last MFC). Greg must have been busy. 8-) My last -stable build was 2/4 and changed up to current shortly after that. Anyone else still have the undying vinum process? Should have been fixed on the 2/14 commit, but it recent creeped back into current (at least for me). At least I can no longer force it to panic, but will be testing the recent current commits again. Jeff Mountin - jeff@mountin.net Systems/Network Administrator FreeBSD - the power to serve To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message