From owner-freebsd-current Mon Jan 29 10: 0:24 2001 Delivered-To: freebsd-current@freebsd.org Received: from mobile.wemm.org (c1315225-a.plstn1.sfba.home.com [65.0.135.147]) by hub.freebsd.org (Postfix) with ESMTP id 791AB37B698 for ; Mon, 29 Jan 2001 10:00:03 -0800 (PST) Received: from netplex.com.au (localhost [127.0.0.1]) by mobile.wemm.org (8.11.1/8.11.1) with ESMTP id f0THxw454602; Mon, 29 Jan 2001 09:59:58 -0800 (PST) (envelope-from peter@netplex.com.au) Message-Id: <200101291759.f0THxw454602@mobile.wemm.org> X-Mailer: exmh version 2.2 06/23/2000 with nmh-1.0.4 To: mjacob@feral.com Cc: Bruce Evans , current@FreeBSD.ORG Subject: Re: config(8) broken In-Reply-To: Date: Mon, 29 Jan 2001 09:59:58 -0800 From: Peter Wemm Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Matthew Jacob wrote: > > *groan* I'm having trouble believing that *config* of all things is now > dependent on time to avoid bugs... This is *one* for the books.... Yep. That's why I disabled it and will replace it with something more robust. In hindsight it was Not A Good Thing. > On Mon, 29 Jan 2001, Peter Wemm wrote: > > > Bruce Evans wrote: > > > On Mon, 29 Jan 2001, Peter Wemm wrote: > > > > > > > Bruce Evans wrote: > > > > > Config now removes almost all headers: > > > > > ... > > > > > This is starting from compile directory populated by a previous versi on > > > > > of config. Starting from scratch, config seems to work for the first > > > > > run. The second run complains about all headers, deletes them all, a nd > > > > > doesn't create any. The third run seems to work... > > > > > > > > What are the chances that you did this on a mount -o noatime file syste m? > > > > > > Very high :-). > > > > Heh. :-) > > > > > I'm surprised the bug doesn't affect fast machines, since config doesn't > > > seem to sleep for >= 1 second to ensure that the atimes advance with > > > the standard mount options. > > > > I was primarily testing it on a dual P3-733 w/ 1G ram which was running > > config nearly instantly. I never saw this problem until I tested it > > on a fresh machine that (by sheer chance) happened to be running noatime (i t > > was a laptop). For what it's worth, I did roll the start time back by one > > second to give a bit more comfort - getting rid of the trays isn't all that > > critical. > > > > For UFS at least, we seem to update the atime instantly but schedule > > writeback to the inode "whenever". > > > > 148# date ; ls -lutT foo1 ; cat foo1 ; ls -lutT foo1; date > > Mon Jan 29 08:18:02 PST 2001 > > -rw-r--r-- 1 root wheel 4 Jan 29 08:16:25 2001 foo1 > > foo > > -rw-r--r-- 1 root wheel 4 Jan 29 08:18:02 2001 foo1 > > Mon Jan 29 08:18:02 PST 2001 > > > > This is all within the same second "tick" and the atime change was visible. > > > > > Bruce > > > > > > Cheers, > > -Peter > > -- > > Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au > > "All of this is for nothing if we don't go to the stars" - JMS/B5 > > > > > > > > To Unsubscribe: send mail to majordomo@FreeBSD.org > > with "unsubscribe freebsd-current" in the body of the message > > > > Cheers, -Peter -- Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message