From owner-freebsd-hackers Tue Feb 6 12:43: 3 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from mailtoaster1.pipeline.ch (mailtoaster1.pipeline.ch [62.48.0.70]) by hub.freebsd.org (Postfix) with SMTP id 9024037B503 for ; Tue, 6 Feb 2001 12:42:39 -0800 (PST) Received: (qmail 14069 invoked from network); 6 Feb 2001 20:39:27 -0000 Received: from unknown (HELO monzoon.net) ([195.134.133.140]) (envelope-sender ) by mailtoaster1.pipeline.ch (qmail-ldap-1.03) with SMTP for ; 6 Feb 2001 20:39:27 -0000 Message-ID: <3A806172.B84B4A69@monzoon.net> Date: Tue, 06 Feb 2001 21:41:22 +0100 From: Andre Oppermann X-Mailer: Mozilla 4.74 [en] (Windows NT 5.0; U) X-Accept-Language: en MIME-Version: 1.0 To: Alfred Perlstein Cc: Rik van Riel , Matt Dillon , Mike Silbersack , Poul-Henning Kamp , Charles Randall , Dan Phoenix , Jos Backus , freebsd-hackers@FreeBSD.ORG Subject: Re: soft updates and qmail (RE: qmail IO problems) References: <3A805E94.8FF4F103@monzoon.net> <20010206123259.U26076@fw.wintelcom.net> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Alfred Perlstein wrote: > > * Andre Oppermann [010206 12:30] wrote: > > Rik van Riel wrote: > > > > > > On Tue, 6 Feb 2001, Andre Oppermann wrote: > > > > > > > But please answer me one question: Is the link() call atomically > > > > in FFS/UFS w or w/o softupdates? Meaning when the call returns > > > > the meta- data is written to stable storage like with fsync()? > > > > > > Since when does `atomic' equal `synchronous' ? > > > > Because otherwise it would not be atomically, would it? > > Softupdates does it atomically but not synchronously. :) > > Basically, the in-memory view of the filesystem != the on-disk > version. > > The update happens atomically with respect to locking in memory, > so running processes never see a non-atomic snapshot of the directory, > but if a crash occurs the disk may be several steps behind the > memory at the time of the crash. Maybe this explains why DJB recommends against the usage of softupdates... But anyway since qmail does fsync() frequently (which is honored by softupdates) then simply the roll-back/forward would kick in. So following this it should be no problem. Maybe DJB recommends against it because softupdates wasn't perfectly mature until recently (and he uses OpenBSD for his developement which AFAIK lags behind FreeBSD in this respect). -- Andre To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message