Date: Mon, 7 Nov 2005 00:11:24 +0300 From: Gleb Smirnoff <glebius@FreeBSD.org> To: Xin LI <delphij@gmail.com> Cc: Yar Tikhiy <yar@comp.chem.msu.su>, freebsd-current@FreeBSD.org, Taras Savchuk <taras.savchuk@gmail.com> Subject: Re: May be a bug in fsck [ after super block crash on 5.4-STABLE ] Message-ID: <20051106211124.GL91530@cell.sick.ru> In-Reply-To: <a78074950511060705r500e0789y786af5d937bf1680@mail.gmail.com> References: <84099c3d0511030325q6d1df92ag77310ff1b03a2d15@mail.gmail.com> <84099c3d0511030425q3592a288he254cb5f97f976b6@mail.gmail.com> <20051106144917.GA81664@comp.chem.msu.su> <a78074950511060705r500e0789y786af5d937bf1680@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Nov 06, 2005 at 11:05:28PM +0800, Xin LI wrote: X> On 11/6/05, Yar Tikhiy <yar@comp.chem.msu.su> wrote: X> > Isn't the type, UFS1 or UFS2, indicated by a magic number in the X> > superblock itself? I used to believe so. If it's true, fsck cannot X> > know the FS type prior to locating a superblock copy. OTOH, with X> > UFS2 having become popular, fsck might try both locations, 32 and 160. X> > Care to file a PR? X> X> That's correct. Fortunately, given that we have some ways to validate X> whether the superblock is valid, it is not too hard to automatically X> detect which type the FS actually is. I think this feature is already present in libufs, since dumpfs(8) can detect UFS1/UFS2 type of filesystem. -- Totus tuus, Glebius. GLEBIUS-RIPN GLEB-RIPE
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20051106211124.GL91530>