Date: Tue, 6 Feb 2001 11:56:12 -0800 (PST) From: Matt Dillon <dillon@earth.backplane.com> To: Andre Oppermann <oppermann@monzoon.net> Cc: Mike Silbersack <silby@silby.com>, Rik van Riel <riel@conectiva.com.br>, Poul-Henning Kamp <phk@critter.freebsd.dk>, Charles Randall <crandall@matchlogic.com>, Dan Phoenix <dphoenix@bravenet.com>, Alfred Perlstein <bright@wintelcom.net>, Jos Backus <josb@cncdsl.com>, freebsd-hackers@FreeBSD.ORG Subject: Re: soft updates and qmail (RE: qmail IO problems) Message-ID: <200102061956.f16JuCb65538@earth.backplane.com> References: <Pine.BSF.4.31.0102061149180.14899-100000@achilles.silby.com> <200102061759.f16Hxv662437@earth.backplane.com> <3A805137.230E0A0D@monzoon.net>
next in thread | previous in thread | raw e-mail | index | archive | help
:
:This information is in fact correct. Have a look at the FreeBSD link(2)
:man page:
:
:LINK(2) FreeBSD System Calls Manual
:LINK(2)
Andre, I think there *might* be a dozen people in the world that
understand UFS/FFS better then I do, but none of them have posted to
this thread.
Believe me when I say that there is no metadata ordering guarentee
in the case of a crash. Yes a standard UFS/FFS will do synchronous
metadata updates for certain operations. No, this does not guarentee
metadata ordering.
There has been talk of providing system calls to allow user programs
to request ordering semantics for certain operations, but nobody has
actually implemented anything. Most of the discussion has been centered
on having calls to guarentee file write ordering between a set of
open descriptors for databases.
As I said, a journaled filesystem can theoretically make metadata
ordering guarentees, but actually doing so creates massive performance
and scaleability issues that aren't apparent until you really start
pounding the filesystem. It *CAN* be done efficiently, but only if
you have a significant amount of non-volatile memory store to hold
the journal. ReiserFS might do it right now, as a side effect, but if
it does it faces serious scaleability issues.
Softupdates can also theoretically order [meta]data, using dependancies.
It is a very difficult problem to solve and it doesn't do it now. All
softupdates does is guarentee filesystem consistency in the case of a
crash and certain guarentees for what will be crash-recoverable when you
do an fsync().
-Matt
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200102061956.f16JuCb65538>