From owner-freebsd-current Wed Mar 3 7: 8:40 1999 Delivered-To: freebsd-current@freebsd.org Received: from fledge.watson.org (FLEDGE.RES.CMU.EDU [128.2.93.229]) by hub.freebsd.org (Postfix) with ESMTP id 178AF14E97 for ; Wed, 3 Mar 1999 07:07:34 -0800 (PST) (envelope-from robert@cyrus.watson.org) Received: from fledge.watson.org (robert@fledge.pr.watson.org [192.0.2.3]) by fledge.watson.org (8.8.8/8.8.8) with SMTP id KAA27508; Wed, 3 Mar 1999 10:04:59 -0500 (EST) (envelope-from robert@cyrus.watson.org) Date: Wed, 3 Mar 1999 10:04:59 -0500 (EST) From: Robert Watson X-Sender: robert@fledge.watson.org Reply-To: Robert Watson To: SXren Schmidt Cc: "Daniel O'Connor" , current@FreeBSD.ORG, geoffr@is.co.za Subject: Re: CALL FOR TESTERS of new ATA/ATAPI driver.. In-Reply-To: <199903030731.IAA86207@freebsd.dk> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Wed, 3 Mar 1999, SXren Schmidt wrote: > It seems Daniel O'Connor wrote: > > > > On 02-Mar-99 SXren Schmidt wrote: > > > Its in the works, together with the tagged queuing some of the > > > newer drives supports. > > Wow! :) > > > > Is there any chance od adding the ability to 'wire' devices a la SCSI? :) > > I'll think about it, but lots of things has higher priority... So, one of my great FreeBSD experiences was when one of the three IDE drives in my server died, yet the system was able to survive its automated weekly boot ten minutes later, and I could restore from a tape backup to another drive. The great thing about this was that I was over 500 miles away logged in via the network performing this recovery. Having the drives automatically renumber themselves on the complete failure of one drive (and it was fairly complete) would have been a disaster. This is especially true when one partition is used as a /tmp partition, and is automatically cleaned at boot :-(. Wiring devices introduce stability by not causing unexpected behavior in the event of failure: rather than having the mounting of partitions occur incorrectly, it merely doesn't mount the failed partitions. In general, I do not like the devices-allocated-in-probe-order behavior: it doesn't take much to reorder two SCSI controllers in a kernel, especially if we start using an automated admin tool to configure kernels at some point. I just don't like that failures can result in very situation-dependent behavior. Inserting a new CDROM device should not renumber my existing devices. This presumably becomes more pressing when one starts doing hot-swap pccard-style devices. If you leave the device plugged in across a reboot, its device name may have changed :-(, or might have gotten stuck ahead of the boot device in the kernel probes. Robert N Watson robert@fledge.watson.org http://www.watson.org/~robert/ PGP key fingerprint: 03 01 DD 8E 15 67 48 73 25 6D 10 FC EC 68 C1 1C Carnegie Mellon University http://www.cmu.edu/ TIS Labs at Network Associates, Inc. http://www.tis.com/ Safeport Network Services http://www.safeport.com/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message