Date: Mon, 7 Dec 2009 17:21:07 -0500 From: jhell <jhell@DataIX.net> To: "James R. Van Artsdalen" <james-freebsd-fs2@jrv.org> Cc: freebsd-fs <freebsd-fs@freebsd.org>, Pawel Jakub Dawidek <pjd@freebsd.org> Subject: Re: ZFS and reordering drives Message-ID: <alpine.BSF.2.00.0912071714560.26308@qvzrafvba.5c.ybpny> In-Reply-To: <4B1D6E66.4000700@jrv.org> References: <20091205170400.GM73250@gremlin.foo.is> <8555674.871260033069220.JavaMail.root@zimbra> <20091205184112.GP73250@gremlin.foo.is> <5da0588e0912051052p25fb743ele098ed9cb9de8fa0@mail.gmail.com> <20091205190641.GQ73250@gremlin.foo.is> <20091205191526.GR73250@gremlin.foo.is> <CA100C20-5F93-4DCE-B576-474DAEC10747@multipart-mixed.com> <20091206122603.GD2339@garage.freebsd.pl> <4B1D6E66.4000700@jrv.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 7 Dec 2009 16:06, james-freebsd-fs2@ wrote: > > I don't see any way to fix this without callers specifying whether a > device is being added to a pool (don't check GUID) or merely attached as > part of a pool (always check GUID). > Prevention of this should IMHO be determined by the type of device being used and method of implementation. Using daN ? devices and the characteristics of those devices could best be handled by glabel before the device is attached to a pool. Maybe there is a way that a device could still be labeled but thus allow to still use the raw device and check what the label is so if that device happens to go missing any device shifts won't effect future operations upon replacing a drive. -- Mon Dec 7 17:14:56 2009 -0500 jhell
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.0912071714560.26308>