From owner-freebsd-hardware@FreeBSD.ORG Fri Dec 21 07:15:44 2007 Return-Path: Delivered-To: freebsd-hardware@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 9EAFB16A41A for ; Fri, 21 Dec 2007 07:15:44 +0000 (UTC) (envelope-from rp_freebsd@mac.com) Received: from smtpoutm.mac.com (smtpoutm.mac.com [17.148.16.66]) by mx1.freebsd.org (Postfix) with ESMTP id 90EF613C44B for ; Fri, 21 Dec 2007 07:15:44 +0000 (UTC) (envelope-from rp_freebsd@mac.com) Received: from webmail039 (webmail039-s [10.13.128.39]) by smtpoutm.mac.com (Xserve/smtpout003/MantshX 4.0) with ESMTP id lBL7FgBk011410; Thu, 20 Dec 2007 23:15:42 -0800 (PST) Date: Thu, 20 Dec 2007 23:15:42 -0800 From: Ravi Pokala To: Ivan Voras , freebsd-hardware@freebsd.org Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Originating-IP: 70.255.1.240 Received: from [70.255.1.240] from webmail.mac.com with HTTP; Thu, 20 Dec 2007 23:15:42 -0800 Cc: Subject: Re: Questions about gmirror X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 21 Dec 2007 07:15:44 -0000 > From: Ivan Voras > Date: Wed, 19 Dec 2007 20:30:37 +0100 > To: > Conversation: Questions about gmirror > Subject: Re: Questions about gmirror > > Ravi Pokala wrote: >> (1) What is the procedure for swapping out a dead drive while the system is >> online? > > If the drive and the controller support hot swapping, remove it, put a > new drive in and do: > > # atacontrol reinit channel > # gmirror forget gm0 > # gmirror insert gm0 /dev/newdrive > > Not many ATA drive+controller combinations are hot-swap-able. > >> (2) Say a drive dies as a result of a power event, and does not respond to >> probing during the subsequent boot. Once I swap out the drive, (how) can I >> get the system to re-probe, recognize the new drive, and start >> reconstruction? > > Exactly the same as above. > I tried that scenario, and it didn't work, which is why I asked. Now, in light of your comment about the controller, I'm looking at dmseg.boot and I see something kind of odd: atapci0: port 0xbc00-0xbc07,0xb880-0xb883,0xb800-0xb807,0xb480-0xb483,0xb400-0xb40f mem 0xfcafe000-0xfcafffff irq 11 at device 14.0 on pci1 ata2: on atapci0 ata3: on atapci0 atapci1: port 0xb080-0xb087,0xb000-0xb003,0xac00-0xac07,0xa880-0xa883,0xa800-0xa80f irq 11 at device 14.1 on pci1 ata4: on atapci1 ata5: on atapci1 atapci2: port 0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xffa0-0xffaf at device 2.1 on pci0 ata0: on atapci2 ata1: on atapci2 ... acd0: CDROM at ata0-master UDMA33 ad4: DMA limited to UDMA33, device found non-ATA66 cable ad4: 476940MB at ata2-master UDMA33 ad6: DMA limited to UDMA33, device found non-ATA66 cable ad6: 476940MB at ata3-master UDMA33 Note the warning about "non-ATA66 cable"... On another machine (for which I have not tried the test, but will after New Year's), using a different controller: atapci0: port 0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xf000-0xf00f at device 31.1 on pci0 ata0: on atapci0 ata1: on atapci0 atapci1: port 0xe900-0xe907,0xea00-0xea03,0xeb00-0xeb07,0xec00-0xec03,0xed00-0xed0f mem 0xd05c3000-0xd05c33ff irq 19 at device 31.2 on pci0 ata2: on atapci1 ata3: on atapci1 acd0: CDROM at ata0-master UDMA33 ad4: 76319MB at ata2-master SATA150 That one recognizes the drive as being SATA, so hopefully it will work better. If it turns out the controller on the first machine can't handle hot-swapping then that's unfortunate, since we bought a few of them. :-/ But on the bright side, we also have a few of the latter as well. Thanks Ivan! --rp