*BSD News Article 62010


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.bhp.com.au!mel.dit.csiro.au!munnari.OZ.AU!news.ecn.uoknor.edu!news.cis.okstate.edu!news.ksu.ksu.edu!news.mid.net!newsfeeder.gi.net!newsfeed.internetmci.com!inet-nntp-gw-1.us.oracle.com!news.caldera.com!news.cc.utah.edu!park.uvsc.edu!not-for-mail
From: mday@park.uvsc.edu (Matt Day)
Newsgroups: comp.unix.bsd.freebsd.misc,comp.os.linux.development.system
Subject: Re: The better (more suitable)Unix?? FreeBSD or Linux
Date: 22 Feb 1996 15:32:40 -0700
Organization: Utah Valley State College, Orem, Utah
Lines: 30
Message-ID: <4giqu8$aqk@park.uvsc.edu>
References: <4gejrb$ogj@floyd.sw.oz.au> <Dn67EK.3nB@pe1chl.ampr.org> <4gilab$97u@park.uvsc.edu>
NNTP-Posting-Host: park.uvsc.edu
Xref: euryale.cc.adfa.oz.au comp.unix.bsd.freebsd.misc:14249 comp.os.linux.development.system:17858

In article <4gilab$97u@park.uvsc.edu> Terry Lambert <terry@lambert.org> writes:
>rob@pe1chl.ampr.org (Rob Janssen) wrote:
>] Don't bother to tell Terry about your experiences, he will just declare
>] your system obsolete and your tools out-of-date.
>] Remember: sync metadata updates only work well on *his* system.
>] On all others, it is just a drag.
>
>Imagine that... I must have a lot of gall to declare obsolete
>a file system implementation that doesn't properly conform to
>the design documents from which it purports to be dervied.
>
>Silly me.
>
>Apparently, not having read the documents, you don't realize
>that it is *impossible* for a conforming implementation to
>result in the errors you want to attribute to UFS -- short
>of a unrecoverable hardware failure.  Do you not believe that
>finite state automatons are deterministic?  I can assure you
>that they are.

It's interesting to note that even FreeBSD, by default, fails to do all
metadata updates safely.  See the ffs_reallocblks() routine in
ufs/ffs/ffs_alloc.c for details (in particular, look for the
"doasyncfree" variable, which is true by default).  So it would appear
that there is some similarity between the Linux ext2fs metadata update
philosophy and the FreeBSD FFS metadata update philosophy after all.
And I don't blame them one bit, either.  Synchronous metadata updates
suck.  I'll be glad if/when the soft updates solution is in.

Matt Day <mday@park.uvsc.edu>