From owner-freebsd-hackers Mon Dec 8 08:16:27 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id IAA17940 for hackers-outgoing; Mon, 8 Dec 1997 08:16:27 -0800 (PST) (envelope-from owner-freebsd-hackers) Received: from firewall.ftf.dk (root@mail.ftf.dk [129.142.64.2]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id IAA17924 for ; Mon, 8 Dec 1997 08:16:05 -0800 (PST) (envelope-from regnauld@deepo.prosa.dk) Received: from mail.prosa.dk ([192.168.100.2]) by firewall.ftf.dk (8.7.6/8.7.3) with ESMTP id SAA31754 for ; Mon, 8 Dec 1997 18:53:44 +0100 Received: from deepo.prosa.dk (deepo.prosa.dk [192.168.100.10]) by mail.prosa.dk (8.8.5/8.8.5/prosa-1.1) with ESMTP id RAA15393 for ; Mon, 8 Dec 1997 17:40:47 +0100 (CET) Received: (from regnauld@localhost) by deepo.prosa.dk (8.8.7/8.8.5/prosa-1.1) id RAA24056; Mon, 8 Dec 1997 17:14:57 +0100 (CET) Message-ID: <19971208171457.56387@deepo.prosa.dk> Date: Mon, 8 Dec 1997 17:14:57 +0100 From: Philippe Regnauld To: hackers@freebsd.org Subject: Question about ccd and disk failure Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.88e X-Operating-System: FreeBSD 2.2.5-RELEASE i386 Sender: owner-freebsd-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk A recent discussion in -current regarding a failing wd0 disk makes me think about something: I have at least two machines (one of them a dedicated dialup box) that use ccd and mirroring (ccd 0x04) on two smallish junk harddisks (i.e.: I'll probably make a floppy out of it one day :-) My thought was: to hell with backups on this box, all the conf. files fit on a floppy, an and if I want to be up quick if/when one of the disks dies, I just remove the bad one, adjust ccd parameters (I've already tried this: using ccd for 'mirroring' on only one of the two remaining disks works great) and reboot. Now my idea is: how tolerant can we be if one of the two disks goes toast ? Meaning, how difficult (implications, 125-gotchas list, etc...) would it be to have ccd just stop writing to a disk after so many errors, and try to failover to the other one only ? Of course we have a problem with / :-( [not subscribed to -hackers] -- -[ Philippe Regnauld / sysadmin / regnauld@deepo.prosa.dk / +55.4N +11.3E ]- "Pluto placed his bad dog at the entrance of Hades to keep the dead IN and the living OUT! The archetypical corporate firewall?" - S. Kelly Bootle, about Cerberus ["MYTHOLOGY", in Marutukku distrib] -