From owner-freebsd-hackers Tue Mar 3 16:11:15 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA25095 for freebsd-hackers-outgoing; Tue, 3 Mar 1998 16:11:15 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from Kitten.mcs.com (Kitten.mcs.com [192.160.127.90]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA24939 for ; Tue, 3 Mar 1998 16:10:30 -0800 (PST) (envelope-from karl@Mars.mcs.net) Received: from Mars.mcs.net (karl@Mars.mcs.net [192.160.127.85]) by Kitten.mcs.com (8.8.7/8.8.2) with ESMTP id SAA16326; Tue, 3 Mar 1998 18:09:59 -0600 (CST) Received: (from karl@localhost) by Mars.mcs.net (8.8.7/8.8.2) id SAA29224; Tue, 3 Mar 1998 18:09:59 -0600 (CST) Message-ID: <19980303180959.19173@mcs.net> Date: Tue, 3 Mar 1998 18:09:59 -0600 From: Karl Denninger To: Wilko Bulte Cc: Julian Elischer , shimon@simon-shapiro.org, hackers@FreeBSD.ORG Subject: Re: SCSI Bus redundancy... References: <34FC66A3.2781E494@whistle.com> <199803032144.WAA03955@yedi.iaf.nl> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.84 In-Reply-To: <199803032144.WAA03955@yedi.iaf.nl>; from Wilko Bulte on Tue, Mar 03, 1998 at 10:44:52PM +0100 Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Tue, Mar 03, 1998 at 10:44:52PM +0100, Wilko Bulte wrote: > Once Upon A Time, When Power Supplies Were Still Powersupplies there were > 2 signals available: AC_OK and DC_OK. Whenever your logic (disk) saw > AC_OK negate, it was time to cleanup. After some time, dependent on how > big your powersupply capacitors were, how loaded the PS was etc you saw > DC_OK negate. Yep. > Current drives can only sense powergood by looking at their DC power inputs. > This is of course lousy, because as soon as you see power drop, you better > park the heads and lock out any write current to the heads. A spiral written > by a retracting head on top of your precious data leaves lots to > be desired from a data integrity standpoint. Well, if you have a big enough cap on the output side of the regulator you can write the rest of the current sector. But these days, trying to get that on the circuit board is basically impossible. > This not even takes into account writing out unflushed cache data, possibly > requiring a seek. And settling time, and acquisition of the servo (along with centering on it), etc etc etc. > Drive write caches are Evil. Every write cache without good battery backup > is Evil. Talk to a DBMS guy about enabling disk write caches. Put sneakers > on and be prepared to run fast... > > But then again, with VM systems that have megabytes worth of unflushed > data the best way to loose your data is to pull the plug from your server > ;-) This is one of the reasons I like the CMD RAID controllers. They have a nice big cache on them (with appropriate SIMMs), but they ALSO have an input for a 6V gelcel battery, and an internal *charging circuit* to manage it. In addition, they have inputs on them to sense UPS health (if you have one). You therefore get three levels of protection: 1) If the UPS goes onto battery, the unit starts "watching" things. 2) If it gets a low power warnings (ie: the "2 minute warning") it flushes the cache and goes into write-through mode. Now you're "safe" if you get screwed. 3) If you get dumped without warning, the battery is there and it will pick up the pieces when power returns. Note that if you have no battery connected or its discharged (the controller is smart enough to know), getting a 2-minute warning flushes the cache and quiesces the controller IMMEDIATELY. These controllers will not operate without either a battery or UPS, and both are (of course) preferred. -- -- Karl Denninger (karl@MCS.Net)| MCSNet - Serving Chicagoland and Wisconsin http://www.mcs.net/ | T1's from $600 monthly to FULL DS-3 Service | NEW! K56Flex support on ALL modems Voice: [+1 312 803-MCS1 x219]| EXCLUSIVE NEW FEATURE ON ALL PERSONAL ACCOUNTS Fax: [+1 312 803-4929] | *SPAMBLOCK* Technology now included at no cost To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message