Date: Mon, 13 Dec 1999 18:57:34 -0800 From: Greg Lehey <grog@lemis.com> To: "T. William Wells" <bill@twwells.com> Cc: freebsd-questions@FreeBSD.ORG Subject: Re: softupdates... the sequel Message-ID: <19991213185734.B358@mojave.worldwide.lemis.com> In-Reply-To: <82ui21$2s8a$1@twwells.com>; from bill@twwells.com on Sat, Dec 11, 1999 at 05:13:00PM -0500 References: <Pine.BSF.4.21.9912110346420.56533-100000@dogma.freebsd-uk.eu.org> <82ui21$2s8a$1@twwells.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Saturday, 11 December 1999 at 17:13:00 -0500, T. William Wells wrote: > In article <Pine.BSF.4.21.9912110346420.56533-100000@dogma.freebsd-uk.eu.org>, > Jonathon McKitrick <jcm@dogma.freebsd-uk.eu.org> wrote: > : (i guess that matters now ;-) does it make any sense to install > : softupdates? Or is it negligible? Or am i sacrificing reliability? > : > : Just so i don't look foolish, here is m guess: > : Yes, i will gain some performance > : No, i will not lose reliability > : > : Correct me if i'm wrong. ;-) > > Softupdates may destroy your file system if you have any I/O > errors. > > Seriously. > > I do have evidence to back this up -- I just recently had a rash > of hard disk failures, in which small portions of the disk started > reporting media failures. Where softupdates was enabled parts of > the disk that were not involved with the hardware failure were > also corrupted. I wish this were adequate. I'm not saying you're wrong, but there are so many different things which could have caused them, and we'd really love to know what the problem is. This isn't so much as a "the problem is in *this* component" problem as a "how do we catch this bug and eliminate it" problem. > In several cases, I ended up with partitions I could not clean > with fsck. All I could do was mount them read-only and copy them > elsewhere. *One* I couldn't even do that -- the root node was > gone. Fortunately, that was just a /var partition. > > I'm not sure what the fix here is. Softupdates *should not* have > this effect. My guess is that what is happening is that > softupdates doesn't handle I/O errors well. The quick hack would > be for softupdates to disable itself as soon as a partition > reports an I/O error.... Since the primary intention of soft updates is to handle I/O errors well, you can believe that we're concerned about this. If you can give us details of what goes wrong, preferably repeatable, we'll be grateful. Yes, this is asking a lot. Greg -- When replying to this message, please copy the original recipients. For more information, see http://www.lemis.com/questions.html Finger grog@lemis.com for PGP public key See complete headers for address and phone numbers To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19991213185734.B358>