From owner-freebsd-fs Sun Mar 24 9:36:25 2002 Delivered-To: freebsd-fs@freebsd.org Received: from fledge.watson.org (fledge.watson.org [204.156.12.50]) by hub.freebsd.org (Postfix) with ESMTP id E753837B419; Sun, 24 Mar 2002 09:36:19 -0800 (PST) Received: from fledge.watson.org (fledge.pr.watson.org [192.0.2.3]) by fledge.watson.org (8.11.6/8.11.6) with SMTP id g2OHaEk76910; Sun, 24 Mar 2002 12:36:15 -0500 (EST) (envelope-from robert@fledge.watson.org) Date: Sun, 24 Mar 2002 12:36:13 -0500 (EST) From: Robert Watson X-Sender: robert@fledge.watson.org To: "Karsten W. Rohrbach" Cc: freebsd-fs@freebsd.org, freebsd-bugs@freebsd.org Subject: Re: Soft update instability with heavy IO and offboard IDE controller In-Reply-To: <20020324182436.B79192@mail.webmonster.de> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Sun, 24 Mar 2002, Karsten W. Rohrbach wrote: > Robert Watson(rwatson@freebsd.org)@2002.03.24 09:22:47 +0000: > > > I have to say that my answer on the ATA write caching is a UPS. :-) > > i hope, this is not meant seriously. an ups keeps your system running in > case of mains power outage, true. it does not help anything in the > following (common) scenarios: > - the power supply blows > - bad power cable > - neutron bombs ;-) > > honestly, judging from the quality of hardware you get in .de the first > two things happen more often than a real power outage, so i'd say an ups > is no viable solution when enabling ata wc. it may give you a warm fuzzy > feeling, but that's about it :-) > > just my EUR 0.02, in case something makes it a FAQ to go into the > handbook... In my area, power grid failure is a real problem, far more likely than power supply failure, unfortunately. Obviously, in the end, it's a tradeoff regarding performance and reliability. If you want real reliability, you have to buy hardware that actually behaves the way the software expects. I don't think anyone would be surprised if a future generation of IDE disks ignored the 'write cache disable' setting to gain performance. Robert N M Watson FreeBSD Core Team, TrustedBSD Project robert@fledge.watson.org NAI Labs, Safeport Network Services To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message