From owner-freebsd-current Sat Sep 19 11:24:52 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id LAA01415 for freebsd-current-outgoing; Sat, 19 Sep 1998 11:24:52 -0700 (PDT) (envelope-from owner-freebsd-current@FreeBSD.ORG) Received: from ns1.yes.no (ns1.yes.no [195.119.24.10]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id LAA01410 for ; Sat, 19 Sep 1998 11:24:47 -0700 (PDT) (envelope-from eivind@bitbox.follo.net) Received: from bitbox.follo.net (bitbox.follo.net [195.204.143.218]) by ns1.yes.no (8.9.1a/8.9.1) with ESMTP id UAA18648; Sat, 19 Sep 1998 20:24:12 +0200 (CEST) Received: (from eivind@localhost) by bitbox.follo.net (8.8.8/8.8.6) id UAA24375; Sat, 19 Sep 1998 20:24:10 +0200 (MET DST) Message-ID: <19980919202409.40703@follo.net> Date: Sat, 19 Sep 1998 20:24:09 +0200 From: Eivind Eklund To: Terry Lambert Cc: Don.Lewis@tsc.tdk.com, current@FreeBSD.ORG Subject: Re: softupdates & fsck References: <19980919123143.36373@follo.net> <199809191719.KAA10028@usr09.primenet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89.1i In-Reply-To: <199809191719.KAA10028@usr09.primenet.com>; from Terry Lambert on Sat, Sep 19, 1998 at 05:19:51PM +0000 Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Sat, Sep 19, 1998 at 05:19:51PM +0000, Terry Lambert wrote: > > > That you are seeing these problems implies that the bwrite ordering > > > guarantees that the driver must provide (i.e., that the blocks will > > > be written in the order requested, and that the writes will not > > > return as completed until the data has been committed to the disk) > > > are not being honored. > > > > Given that most drives don't honour these guarantees [1] it may happen > > even without a problem with the driver. > > > > [1] This marks the point where somebody comes runing, waving standards > > documents and becoming more and more red in the face, while I say > > "Yes, I know they say the drives are supposed to - but in fact, the > > drives don't actually *do* what they're supposed to." > > They do if you set their options correctly and insure a holdup > time after power failure, during which you will not engage in > scheduling new writes. Eh? Without an UPS, I can't do that. The problem is drives that don't honour the immediate-write flag, returning before data is committed to disk anyway (done to get better benchmarks). I think there also is some problems with drives that will re-order requests no matter what constraints you give it :-( BTW: I didn't mean that this applied in this particular case, just that this may become a problem when soft updates become fsck-less. Eivind. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message