From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 02:13:34 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 7B10916A4CE for ; Sun, 24 Apr 2005 02:13:34 +0000 (GMT) Received: from gromit.dlib.vt.edu (gromit.dlib.vt.edu [128.173.49.29]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0F8A443D41 for ; Sun, 24 Apr 2005 02:13:34 +0000 (GMT) (envelope-from paul@gromit.dlib.vt.edu) Received: from zappa.Chelsea-Ct.Org (pool-70-110-10-69.roa.east.verizon.net [70.110.10.69]) by gromit.dlib.vt.edu (8.13.3/8.13.3) with ESMTP id j3O2DVlO019520 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Sat, 23 Apr 2005 22:13:32 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) Received: from zappa.Chelsea-Ct.Org (localhost.Chelsea-Ct.Org [127.0.0.1]) by zappa.Chelsea-Ct.Org (8.13.3/8.13.3) with ESMTP id j3O2DO04071963 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Sat, 23 Apr 2005 22:13:26 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) Received: (from paul@localhost) by zappa.Chelsea-Ct.Org (8.13.3/8.13.3/Submit) id j3O2DMl2071962 for freebsd-geom@freebsd.org; Sat, 23 Apr 2005 22:13:23 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) X-Authentication-Warning: zappa.Chelsea-Ct.Org: paul set sender to paul@gromit.dlib.vt.edu using -f From: Paul Mather To: freebsd-geom@freebsd.org Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Sat, 23 Apr 2005 22:13:21 -0400 Message-Id: <1114308801.71938.2.camel@zappa.Chelsea-Ct.Org> Mime-Version: 1.0 X-Mailer: Evolution 2.2.2 FreeBSD GNOME Team Port Subject: Is there a "disconnected" state for geom_mirror providers? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 02:13:34 -0000 Sadly, the "TIMEOUT - WRITE_DMA"-induced disk disconnections have returned on my -CURRENT system since I upgraded to ATA Mk.III. :-( However, I've noticed that when a drive is marked as failed and the device detached, the provider also disappears from the geom_mirror it is part of, instead of being marked as a "stale" or "disconnected" or "missing" component of the remaining mirror components. Is this the correct behaviour? In the latest failure to occur, ad0 was detached: ad0: TIMEOUT - WRITE_DMA retrying (2 retries left) LBA=49981679 ad0: FAILURE - device detached subdisk0: detached ad0: detached GEOM_MIRROR: Cannot update metadata on disk ad0 (error=5). GEOM_MIRROR: Cannot update metadata on disk ad0 (error=6). GEOM_MIRROR: Device raid1: provider ad0 disconnected. GEOM_MIRROR: Request failed (error=6). ad0[WRITE(offset=3847741440, length=16384)] I performed an "atacontrol detach 0" followed by an "atacontrol attach 0" to "re-discover" the "failed" ad0 as part of the existing geom_mirror. This yielded the following: acd0: detached (cd0:ata0:0:1:0): lost device (cd0:ata0:0:1:0): removing device entry atapicam0: detached stray irq14 ad0: 24405MB at ata0-master UDMA33 GEOM_MIRROR: Component ad0 (device raid1) broken, skipping. GEOM_MIRROR: Cannot add disk ad0 to raid1 (error=22). acd0: DVDR at ata0-slave UDMA33 cd0 at ata0 bus 0 target 1 lun 0 cd0: Removable CD-ROM SCSI-0 device cd0: 33.000MB/s transfers cd0: cd present [1 x 2048 byte records] The provider ad0 did not show up as a "stale" provider of my "raid1" mirror (from which it had disappeared when it was detached due to the "TIMEOUT - WRITE_DMA" failure). I had to do a "gmirror forget raid1" before a "gmirror insert raid1 ad0" would allow me to re-insert it so I could perform a "gmirror rebuild raid1 ad0" to kick off synchronisation. What is the definition of a "broken" component? What is the difference between a "stale" and a "broken" component? If I were to detach and remove a hot-plug geom_mirror component and subsequently re-attach it, will the component be considered "stale" or "broken?" This is not a major inconvenience (well, the return of the "TIMEOUT - WRITE_DMA" errors are:), but I was just wondering why my failed providers disappear now as opposed to being marked as stale as happened in the past. BTW, my system is a fairly recent -CURRENT: FreeBSD 6.0-CURRENT #0: Mon Apr 18 12:25:24 EDT 2005. Cheers, Paul. -- e-mail: paul@gromit.dlib.vt.edu "Without music to decorate it, time is just a bunch of boring production deadlines or dates by which bills must be paid." --- Frank Vincent Zappa From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 02:24:43 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id BF88F16A4CE; Sun, 24 Apr 2005 02:24:43 +0000 (GMT) Received: from mx.toxahost.ru (ns.toxahost.ru [62.89.204.62]) by mx1.FreeBSD.org (Postfix) with ESMTP id E1CBF43D4C; Sun, 24 Apr 2005 02:24:42 +0000 (GMT) (envelope-from toxa@toxahost.ru) Received: from localhost (laptoxa.toxa.lan [192.168.1.3]) by mx.toxahost.ru (Toxa) with ESMTP id A310D76; Sun, 24 Apr 2005 06:25:02 +0400 (MSD) Date: Sun, 24 Apr 2005 06:24:34 +0400 From: Toxa To: FreeBSD-CURRENT X-Comment-To: "Anton A. Karpov" Message-ID: <20050424022434.GA3384@laptoxa.toxa.lan> Mail-Followup-To: FreeBSD-CURRENT , freebsd-geom@freebsd.org Mime-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Disposition: inline User-Agent: Outluck Express 1.5.6i for MS-DOS 6.22-SMP X-Mailer: See User-Agent above :) X-Operating-System: MS-DOS 6.22-CURRENT on Sony VAIO laptop X-PGP-Public-Key: http://toxahost.ru/gpg/pubkey.asc X-Useless-Header: Do Androids Dream of Electric Sheep? cc: freebsd-geom@freebsd.org Subject: gmirror: Not all disks connected X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 02:24:43 -0000 I've installed 5.4-rc3 recently and decided to play with geom gmirror. I have two fresh SAMSUNG SP0411N/TW100-11, so I've put it together and followed steps described in http://people.freebsd.org/~rse/mirror/. Everything was ok, but when I rebooted with "the final two-disk GEOM mirror setup", I've noticed what it's probably still one disk in gm0: Timecounter "TSC" frequency 1817908512 Hz quality 800 Timecounters tick every 10.000 msec ad0: 38203MB [77619/16/63] at ata0-master UDMA100 ad1: 38204MB [77622/16/63] at ata0-slave UDMA100 GEOM_MIRROR: Device gm0 created (id=2404813840). GEOM_MIRROR: Device gm0: provider ad1 detected. GEOM_MIRROR: Force device gm0 start due to timeout. GEOM_MIRROR: Device gm0: provider ad1 activated. GEOM_MIRROR: Device gm0: provider mirror/gm0 launched. Mounting root from ufs:/dev/mirror/gm0s1a Well, maybe I've lost last changes ang gmirror still lives with ad1 only? cymbal# gmirror list Geom name: gm0 State: DEGRADED Components: 2 Balance: round-robin Slice: 4096 Flags: NONE GenID: 0 SyncID: 3 ID: 2404813840 Providers: 1. Name: mirror/gm0 Mediasize: 40060403200 (37G) Sectorsize: 512 Mode: r7w7e2 Consumers: 1. Name: ad1 Mediasize: 40060403712 (37G) Sectorsize: 512 Mode: r1w1e1 State: ACTIVE Priority: 0 Flags: DIRTY GenID: 0 SyncID: 3 ID: 3247857729 When I tried to repeat last steps, I got message "Not all disks connected". cymbal# gmirror configure -a gm0 Not all disks connected. cymbal# gmirror status Name Status Components mirror/gm0 DEGRADED ad1 It seems like gmirror think I've lost ad0, but it's ok. I've tried to play with gmirror utility but in most cases I've got "Not all disk connected" message. Any help will be very appreciated. Thanks in advance. From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 09:41:50 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5595A16A4CE for ; Sun, 24 Apr 2005 09:41:50 +0000 (GMT) Received: from darkness.comp.waw.pl (darkness.comp.waw.pl [195.117.238.136]) by mx1.FreeBSD.org (Postfix) with ESMTP id B082143D5C for ; Sun, 24 Apr 2005 09:41:49 +0000 (GMT) (envelope-from pjd@darkness.comp.waw.pl) Received: by darkness.comp.waw.pl (Postfix, from userid 1009) id 2F0C0ACAEE; Sun, 24 Apr 2005 11:41:48 +0200 (CEST) Date: Sun, 24 Apr 2005 11:41:48 +0200 From: Pawel Jakub Dawidek To: Paul Mather Message-ID: <20050424094148.GZ837@darkness.comp.waw.pl> References: <1114308801.71938.2.camel@zappa.Chelsea-Ct.Org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="IrYWVTOpV5r4MFme" Content-Disposition: inline In-Reply-To: <1114308801.71938.2.camel@zappa.Chelsea-Ct.Org> User-Agent: Mutt/1.4.2i X-PGP-Key-URL: http://people.freebsd.org/~pjd/pjd.asc X-OS: FreeBSD 5.2.1-RC2 i386 cc: freebsd-geom@freebsd.org Subject: Re: Is there a "disconnected" state for geom_mirror providers? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 09:41:50 -0000 --IrYWVTOpV5r4MFme Content-Type: text/plain; charset=iso-8859-2 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sat, Apr 23, 2005 at 10:13:21PM -0400, Paul Mather wrote: +> The provider ad0 did not show up as a "stale" provider of my "raid1" +> mirror (from which it had disappeared when it was detached due to the +> "TIMEOUT - WRITE_DMA" failure). I had to do a "gmirror forget raid1" +> before a "gmirror insert raid1 ad0" would allow me to re-insert it so I +> could perform a "gmirror rebuild raid1 ad0" to kick off synchronisation. +>=20 +> What is the definition of a "broken" component? What is the difference +> between a "stale" and a "broken" component? If gmirror gets an error for READ or WRITE operation, it assumes provider is broken. This is very important - if it will be marked only as stale, it will be connected, resynchronization will start, but because there was an error on provider, it probably will be disconnected again and we have endless loop. Stale provider is when it is hot-plug and you remove it; when you use 'deactivate' command to disconnect it; when it doesn't show up on mirror start, but later. The rule is simple: when an error was returned on I/O operation, provider is marked as broken. +> If I were to detach and remove a hot-plug geom_mirror component and +> subsequently re-attach it, will the component be considered "stale" or +> "broken?" Stale. --=20 Pawel Jakub Dawidek http://www.wheel.pl pjd@FreeBSD.org http://www.FreeBSD.org FreeBSD committer Am I Evil? Yes, I Am! --IrYWVTOpV5r4MFme Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFCa2ncForvXbEpPzQRAi5GAKDgw2HiZ1Sh9tNAcRi/PIfR+nQA2wCePTB1 iFNpfEjw+tI3LwzvViJrEf0= =FeO9 -----END PGP SIGNATURE----- --IrYWVTOpV5r4MFme-- From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 09:43:45 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 6320E16A4CE; Sun, 24 Apr 2005 09:43:45 +0000 (GMT) Received: from darkness.comp.waw.pl (darkness.comp.waw.pl [195.117.238.136]) by mx1.FreeBSD.org (Postfix) with ESMTP id DC61043D48; Sun, 24 Apr 2005 09:43:44 +0000 (GMT) (envelope-from pjd@darkness.comp.waw.pl) Received: by darkness.comp.waw.pl (Postfix, from userid 1009) id B038DACAEE; Sun, 24 Apr 2005 11:43:43 +0200 (CEST) Date: Sun, 24 Apr 2005 11:43:43 +0200 From: Pawel Jakub Dawidek To: FreeBSD-CURRENT , freebsd-geom@freebsd.org Message-ID: <20050424094343.GA837@darkness.comp.waw.pl> References: <20050424022434.GA3384@laptoxa.toxa.lan> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="c5jIKqLJqsIKqNdf" Content-Disposition: inline In-Reply-To: <20050424022434.GA3384@laptoxa.toxa.lan> User-Agent: Mutt/1.4.2i X-PGP-Key-URL: http://people.freebsd.org/~pjd/pjd.asc X-OS: FreeBSD 5.2.1-RC2 i386 Subject: Re: gmirror: Not all disks connected X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 09:43:45 -0000 --c5jIKqLJqsIKqNdf Content-Type: text/plain; charset=iso-8859-2 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sun, Apr 24, 2005 at 06:24:34AM +0400, Toxa wrote: +> I've installed 5.4-rc3 recently and decided to play with geom gmirror. I +> have two fresh SAMSUNG SP0411N/TW100-11, so I've put it together and +> followed steps described in http://people.freebsd.org/~rse/mirror/. +> Everything was ok, but when I rebooted with "the final two-disk GEOM +> mirror setup", I've noticed what it's probably still one disk in gm0: +>=20 +> Timecounter "TSC" frequency 1817908512 Hz quality 800 +> Timecounters tick every 10.000 msec +> ad0: 38203MB [77619/16/63] at ata0-master +> UDMA100 +> ad1: 38204MB [77622/16/63] at ata0-slave +> UDMA100 +> GEOM_MIRROR: Device gm0 created (id=3D2404813840). +> GEOM_MIRROR: Device gm0: provider ad1 detected. +> GEOM_MIRROR: Force device gm0 start due to timeout. +> GEOM_MIRROR: Device gm0: provider ad1 activated. +> GEOM_MIRROR: Device gm0: provider mirror/gm0 launched. +> Mounting root from ufs:/dev/mirror/gm0s1a +>=20 +>=20 +> Well, maybe I've lost last changes ang gmirror still lives with ad1 only= ?=20 +>=20 +> cymbal# gmirror list +> Geom name: gm0 +> State: DEGRADED +> Components: 2 +> Balance: round-robin +> Slice: 4096 +> Flags: NONE +> GenID: 0 +> SyncID: 3 +> ID: 2404813840 +> Providers: +> 1. Name: mirror/gm0 +> Mediasize: 40060403200 (37G) +> Sectorsize: 512 +> Mode: r7w7e2 +> Consumers: +> 1. Name: ad1 +> Mediasize: 40060403712 (37G) +> Sectorsize: 512 +> Mode: r1w1e1 +> State: ACTIVE +> Priority: 0 +> Flags: DIRTY +> GenID: 0 +> SyncID: 3 +> ID: 3247857729 +>=20 +>=20 +> When I tried to repeat last steps, I got message "Not all disks connecte= d". +>=20 +> cymbal# gmirror configure -a gm0 +> Not all disks connected. +> cymbal# gmirror status +> Name Status Components +> mirror/gm0 DEGRADED ad1 +>=20 +>=20 +> It seems like gmirror think I've lost ad0, but it's ok. +> I've tried to play with gmirror utility but in most cases I've got "Not +> all disk connected" message. Please read description of 'forget' subcommand. --=20 Pawel Jakub Dawidek http://www.wheel.pl pjd@FreeBSD.org http://www.FreeBSD.org FreeBSD committer Am I Evil? Yes, I Am! --c5jIKqLJqsIKqNdf Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFCa2pPForvXbEpPzQRAtl3AJ93IKlXe21+NQHadFtFNW4dwmVjEQCePswU SmxIVuYaHJR/38/6mpe3pOk= =Ewse -----END PGP SIGNATURE----- --c5jIKqLJqsIKqNdf-- From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 16:32:02 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3E2DF16A4CE; Sun, 24 Apr 2005 16:32:02 +0000 (GMT) Received: from gromit.dlib.vt.edu (gromit.dlib.vt.edu [128.173.49.29]) by mx1.FreeBSD.org (Postfix) with ESMTP id C5C2E43D41; Sun, 24 Apr 2005 16:32:01 +0000 (GMT) (envelope-from paul@gromit.dlib.vt.edu) Received: from zappa.Chelsea-Ct.Org (pool-70-110-10-69.roa.east.verizon.net [70.110.10.69]) by gromit.dlib.vt.edu (8.13.3/8.13.3) with ESMTP id j3OGVxMg032981 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Sun, 24 Apr 2005 12:32:00 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) Received: from zappa.Chelsea-Ct.Org (localhost.Chelsea-Ct.Org [127.0.0.1]) by zappa.Chelsea-Ct.Org (8.13.3/8.13.3) with ESMTP id j3OGVssg077591 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Sun, 24 Apr 2005 12:31:54 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) Received: (from paul@localhost) by zappa.Chelsea-Ct.Org (8.13.3/8.13.3/Submit) id j3OGVsdH077590; Sun, 24 Apr 2005 12:31:54 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) X-Authentication-Warning: zappa.Chelsea-Ct.Org: paul set sender to paul@gromit.dlib.vt.edu using -f From: Paul Mather To: Pawel Jakub Dawidek In-Reply-To: <20050424094148.GZ837@darkness.comp.waw.pl> References: <1114308801.71938.2.camel@zappa.Chelsea-Ct.Org> <20050424094148.GZ837@darkness.comp.waw.pl> Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Sun, 24 Apr 2005 12:31:53 -0400 Message-Id: <1114360313.77313.14.camel@zappa.Chelsea-Ct.Org> Mime-Version: 1.0 X-Mailer: Evolution 2.2.2 FreeBSD GNOME Team Port cc: freebsd-geom@freebsd.org Subject: Re: Is there a "disconnected" state for geom_mirror providers? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 16:32:02 -0000 On Sun, 2005-04-24 at 11:41 +0200, Pawel Jakub Dawidek wrote: > On Sat, Apr 23, 2005 at 10:13:21PM -0400, Paul Mather wrote: > +> The provider ad0 did not show up as a "stale" provider of my "raid1" > +> mirror (from which it had disappeared when it was detached due to the > +> "TIMEOUT - WRITE_DMA" failure). I had to do a "gmirror forget raid1" > +> before a "gmirror insert raid1 ad0" would allow me to re-insert it so I > +> could perform a "gmirror rebuild raid1 ad0" to kick off synchronisation. > +> > +> What is the definition of a "broken" component? What is the difference > +> between a "stale" and a "broken" component? > > If gmirror gets an error for READ or WRITE operation, it assumes provider > is broken. This is very important - if it will be marked only as stale, > it will be connected, resynchronization will start, but because there > was an error on provider, it probably will be disconnected again and we > have endless loop. I guess it depends on what caused the disconnection in the first place. If it was a READ of a bad sector, it could be that subsequent resynchronisation will force a block reallocation of the bad block and the drive will no longer be "broken." > Stale provider is when it is hot-plug and you remove it; when you use > 'deactivate' command to disconnect it; when it doesn't show up on mirror > start, but later. > > The rule is simple: when an error was returned on I/O operation, provider > is marked as broken. Thanks for the clarification. That makes sense. I just need to remember "gmirror forget" before I attempt to add back in the disk in my "TIMEOUT - WRITE_DMA" not-really-broken broken disk case. :-) The shame about it being deleted from the mirror as opposed to marked as "broken" is you lose info (shown in "gmirror list") about the broken component priority, etc., which is useful for when you add a replacement device (or re-add the same one, as in my case). If you marked a component as "broken" (but still listed as part of the mirror), you could add a "-f" option to "gmirror rebuild" to force rebuilding onto it a la RAIDframe. :-) Cheers, Paul. -- e-mail: paul@gromit.dlib.vt.edu "Without music to decorate it, time is just a bunch of boring production deadlines or dates by which bills must be paid." --- Frank Vincent Zappa From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 17:04:18 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 381F316A4CE for ; Sun, 24 Apr 2005 17:04:18 +0000 (GMT) Received: from darkness.comp.waw.pl (darkness.comp.waw.pl [195.117.238.136]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9698043D55 for ; Sun, 24 Apr 2005 17:04:17 +0000 (GMT) (envelope-from pjd@darkness.comp.waw.pl) Received: by darkness.comp.waw.pl (Postfix, from userid 1009) id 9B769ACBCB; Sun, 24 Apr 2005 19:04:15 +0200 (CEST) Date: Sun, 24 Apr 2005 19:04:15 +0200 From: Pawel Jakub Dawidek To: Paul Mather Message-ID: <20050424170415.GC837@darkness.comp.waw.pl> References: <1114308801.71938.2.camel@zappa.Chelsea-Ct.Org> <20050424094148.GZ837@darkness.comp.waw.pl> <1114360313.77313.14.camel@zappa.Chelsea-Ct.Org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="GLdS9qjAGFrs7Ts6" Content-Disposition: inline In-Reply-To: <1114360313.77313.14.camel@zappa.Chelsea-Ct.Org> User-Agent: Mutt/1.4.2i X-PGP-Key-URL: http://people.freebsd.org/~pjd/pjd.asc X-OS: FreeBSD 5.2.1-RC2 i386 cc: freebsd-geom@freebsd.org Subject: Re: Is there a "disconnected" state for geom_mirror providers? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 17:04:18 -0000 --GLdS9qjAGFrs7Ts6 Content-Type: text/plain; charset=iso-8859-2 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sun, Apr 24, 2005 at 12:31:53PM -0400, Paul Mather wrote: +> > If gmirror gets an error for READ or WRITE operation, it assumes provi= der +> > is broken. This is very important - if it will be marked only as stale, +> > it will be connected, resynchronization will start, but because there +> > was an error on provider, it probably will be disconnected again and we +> > have endless loop. +>=20 +> I guess it depends on what caused the disconnection in the first place. +> If it was a READ of a bad sector, it could be that subsequent +> resynchronisation will force a block reallocation of the bad block and +> the drive will no longer be "broken." So you want me to count number of failures of every sector and mark component as broken if I've 2 failures related to the same sector or something like that?:) +> > Stale provider is when it is hot-plug and you remove it; when you use +> > 'deactivate' command to disconnect it; when it doesn't show up on mirr= or +> > start, but later. +> >=20 +> > The rule is simple: when an error was returned on I/O operation, provi= der +> > is marked as broken. +>=20 +> Thanks for the clarification. That makes sense. I just need to +> remember "gmirror forget" before I attempt to add back in the disk in my +> "TIMEOUT - WRITE_DMA" not-really-broken broken disk case. :-) If reallocation happens here, there should be no I/O error visible for gmirror. +> The shame about it being deleted from the mirror as opposed to marked as +> "broken" is you lose info (shown in "gmirror list") about the broken +> component priority, etc., which is useful for when you add a replacement +> device (or re-add the same one, as in my case). You can use 'gmirror dump /dev/'. +> If you marked a component as "broken" (but still listed as part of the +> mirror), you could add a "-f" option to "gmirror rebuild" to force +> rebuilding onto it a la RAIDframe. :-) This is not so simple. I don't store any info on broken component, that it is broken, because e.g. bad sector could be the sector with metadata. Other components are informed that something wrong is going on. How one can remove such broken component for good? Let's say you was able to read metadata from the component, but you cannot write there any more. How you can easily replace this component? This complicates things a lot and I don't need more complications if I want gmirror to stay reliable (which I hope it is now). --=20 Pawel Jakub Dawidek http://www.wheel.pl pjd@FreeBSD.org http://www.FreeBSD.org FreeBSD committer Am I Evil? Yes, I Am! --GLdS9qjAGFrs7Ts6 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD4DBQFCa9GPForvXbEpPzQRAiaEAJ9ijXlSdmaSEiVZmzmlMG/Qpv+QsQCY0QFF mAViO3NgqYE8BH+7ojxc4Q== =fVKR -----END PGP SIGNATURE----- --GLdS9qjAGFrs7Ts6-- From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 17:49:58 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 75A9E16A4CE; Sun, 24 Apr 2005 17:49:58 +0000 (GMT) Received: from gromit.dlib.vt.edu (gromit.dlib.vt.edu [128.173.49.29]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0D72443D46; Sun, 24 Apr 2005 17:49:58 +0000 (GMT) (envelope-from paul@gromit.dlib.vt.edu) Received: from zappa.Chelsea-Ct.Org (pool-70-110-10-69.roa.east.verizon.net [70.110.10.69]) by gromit.dlib.vt.edu (8.13.3/8.13.3) with ESMTP id j3OHnuKS033103 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Sun, 24 Apr 2005 13:49:57 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) Received: from zappa.Chelsea-Ct.Org (localhost.Chelsea-Ct.Org [127.0.0.1]) by zappa.Chelsea-Ct.Org (8.13.3/8.13.3) with ESMTP id j3OHnoeX077847 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Sun, 24 Apr 2005 13:49:50 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) Received: (from paul@localhost) by zappa.Chelsea-Ct.Org (8.13.3/8.13.3/Submit) id j3OHnoBV077846; Sun, 24 Apr 2005 13:49:50 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) X-Authentication-Warning: zappa.Chelsea-Ct.Org: paul set sender to paul@gromit.dlib.vt.edu using -f From: Paul Mather To: Pawel Jakub Dawidek In-Reply-To: <20050424170415.GC837@darkness.comp.waw.pl> References: <1114308801.71938.2.camel@zappa.Chelsea-Ct.Org> <20050424094148.GZ837@darkness.comp.waw.pl> <1114360313.77313.14.camel@zappa.Chelsea-Ct.Org> <20050424170415.GC837@darkness.comp.waw.pl> Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Sun, 24 Apr 2005 13:49:49 -0400 Message-Id: <1114364989.77743.14.camel@zappa.Chelsea-Ct.Org> Mime-Version: 1.0 X-Mailer: Evolution 2.2.2 FreeBSD GNOME Team Port cc: freebsd-geom@freebsd.org Subject: Re: Is there a "disconnected" state for geom_mirror providers? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 17:49:58 -0000 On Sun, 2005-04-24 at 19:04 +0200, Pawel Jakub Dawidek wrote: > On Sun, Apr 24, 2005 at 12:31:53PM -0400, Paul Mather wrote: > +> > If gmirror gets an error for READ or WRITE operation, it assumes provider > +> > is broken. This is very important - if it will be marked only as stale, > +> > it will be connected, resynchronization will start, but because there > +> > was an error on provider, it probably will be disconnected again and we > +> > have endless loop. > +> > +> I guess it depends on what caused the disconnection in the first place. > +> If it was a READ of a bad sector, it could be that subsequent > +> resynchronisation will force a block reallocation of the bad block and > +> the drive will no longer be "broken." > > So you want me to count number of failures of every sector and mark > component as broken if I've 2 failures related to the same sector or > something like that?:) No, I was just pointing out that the "endless loop" scenario you gave might well not hold for certain common classes of read-induced failures. > +> Thanks for the clarification. That makes sense. I just need to > +> remember "gmirror forget" before I attempt to add back in the disk in my > +> "TIMEOUT - WRITE_DMA" not-really-broken broken disk case. :-) > > If reallocation happens here, there should be no I/O error visible for > gmirror. The "TIMEOUT - WRITE_DMA" failure does not involve any bad blocks; it's a persistent problem with the ATA code (hopefully fixed recently:), not the hardware. > +> The shame about it being deleted from the mirror as opposed to marked as > +> "broken" is you lose info (shown in "gmirror list") about the broken > +> component priority, etc., which is useful for when you add a replacement > +> device (or re-add the same one, as in my case). > > You can use 'gmirror dump /dev/'. Thanks! I guess I missed that in the man page. > +> If you marked a component as "broken" (but still listed as part of the > +> mirror), you could add a "-f" option to "gmirror rebuild" to force > +> rebuilding onto it a la RAIDframe. :-) > > This is not so simple. I don't store any info on broken component, that it > is broken, because e.g. bad sector could be the sector with metadata. > Other components are informed that something wrong is going on. > How one can remove such broken component for good? Let's say you was able > to read metadata from the component, but you cannot write there any more. > How you can easily replace this component? If "gmirror rebuild -f" was used, it would imply autosynchronisation was turned off. So, if you had real hardware problems with the provider, it would remain broken because the rebuild would fail, too, with some hardware error. Eventually, as an operator, you'd get the hint that the provider really had lasting problems, and replace it with something that really worked. ;-) > This complicates things a lot and I don't need more complications if I > want gmirror to stay reliable (which I hope it is now). I agree, as as there are ways around the issue via the gmirror command set you've provided to us, it's probably wisest to keep things the way they are. Thanks for geom_mirror (and geom_stripe et al.)! Cheers, Paul. -- e-mail: paul@gromit.dlib.vt.edu "Without music to decorate it, time is just a bunch of boring production deadlines or dates by which bills must be paid." --- Frank Vincent Zappa From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 18:09:08 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id AA1EB16A4CE for ; Sun, 24 Apr 2005 18:09:08 +0000 (GMT) Received: from darkness.comp.waw.pl (darkness.comp.waw.pl [195.117.238.136]) by mx1.FreeBSD.org (Postfix) with ESMTP id CB6E543D68 for ; Sun, 24 Apr 2005 18:09:07 +0000 (GMT) (envelope-from pjd@darkness.comp.waw.pl) Received: by darkness.comp.waw.pl (Postfix, from userid 1009) id 0E16BACAFB; Sun, 24 Apr 2005 20:09:06 +0200 (CEST) Date: Sun, 24 Apr 2005 20:09:06 +0200 From: Pawel Jakub Dawidek To: Paul Mather Message-ID: <20050424180906.GE837@darkness.comp.waw.pl> References: <1114308801.71938.2.camel@zappa.Chelsea-Ct.Org> <20050424094148.GZ837@darkness.comp.waw.pl> <1114360313.77313.14.camel@zappa.Chelsea-Ct.Org> <20050424170415.GC837@darkness.comp.waw.pl> <1114364989.77743.14.camel@zappa.Chelsea-Ct.Org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="pjZlUIsMYcU39AAc" Content-Disposition: inline In-Reply-To: <1114364989.77743.14.camel@zappa.Chelsea-Ct.Org> User-Agent: Mutt/1.4.2i X-PGP-Key-URL: http://people.freebsd.org/~pjd/pjd.asc X-OS: FreeBSD 5.2.1-RC2 i386 cc: freebsd-geom@freebsd.org Subject: Re: Is there a "disconnected" state for geom_mirror providers? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 18:09:08 -0000 --pjZlUIsMYcU39AAc Content-Type: text/plain; charset=iso-8859-2 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sun, Apr 24, 2005 at 01:49:49PM -0400, Paul Mather wrote: +> > So you want me to count number of failures of every sector and mark +> > component as broken if I've 2 failures related to the same sector or +> > something like that?:) +>=20 +> No, I was just pointing out that the "endless loop" scenario you gave +> might well not hold for certain common classes of read-induced failures. I've no way to detect what kind of failure EIO is, that's why I need a general solution. +> > +> The shame about it being deleted from the mirror as opposed to mark= ed as +> > +> "broken" is you lose info (shown in "gmirror list") about the broken +> > +> component priority, etc., which is useful for when you add a replac= ement +> > +> device (or re-add the same one, as in my case). +> >=20 +> > You can use 'gmirror dump /dev/'. +>=20 +> Thanks! I guess I missed that in the man page. Maybe because it wasn't documented:) I missed this command in gmirror manual page, but it is fixed in HEAD and RELENG_5 already. +> > +> If you marked a component as "broken" (but still listed as part of = the +> > +> mirror), you could add a "-f" option to "gmirror rebuild" to force +> > +> rebuilding onto it a la RAIDframe. :-) +> >=20 +> > This is not so simple. I don't store any info on broken component, tha= t it +> > is broken, because e.g. bad sector could be the sector with metadata. +> > Other components are informed that something wrong is going on. +> > How one can remove such broken component for good? Let's say you was a= ble +> > to read metadata from the component, but you cannot write there any mo= re. +> > How you can easily replace this component? +>=20 +> If "gmirror rebuild -f" was used, it would imply autosynchronisation was +> turned off. So, if you had real hardware problems with the provider, it +> would remain broken because the rebuild would fail, too, with some +> hardware error. Eventually, as an operator, you'd get the hint that the +> provider really had lasting problems, and replace it with something that +> really worked. ;-) Imagine something like this: - da1 is broken, but still connected to the mirror. - gmirror rebuild -f da1 doesn't work. - gmirror remove da1 also doesn't work, because we've an error when updating metadata. - gmirror insert da4 (spare disk) also doesn't work, because I cannot update metadata on all components. etc. --=20 Pawel Jakub Dawidek http://www.wheel.pl pjd@FreeBSD.org http://www.FreeBSD.org FreeBSD committer Am I Evil? Yes, I Am! --pjZlUIsMYcU39AAc Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFCa+DCForvXbEpPzQRApdvAKDydK/xQfjjeO6Gyh24kLJHv+fNuQCfXMg/ Hb+Wf0UsRVXj5ICiPvM2aXY= =PxRX -----END PGP SIGNATURE----- --pjZlUIsMYcU39AAc-- From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 19:09:17 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id CB19F16A4CE; Sun, 24 Apr 2005 19:09:17 +0000 (GMT) Received: from mx.toxahost.ru (ns.toxahost.ru [62.89.204.62]) by mx1.FreeBSD.org (Postfix) with ESMTP id 685C543D54; Sun, 24 Apr 2005 19:09:17 +0000 (GMT) (envelope-from toxa@toxahost.ru) Received: from localhost (laptoxa.toxa.lan [192.168.1.3]) by mx.toxahost.ru (Toxa) with ESMTP id 1FA9B122; Sun, 24 Apr 2005 23:09:16 +0400 (MSD) Date: Sun, 24 Apr 2005 23:09:12 +0400 From: Toxa To: freebsd-current@freebsd.org, freebsd-geom@freebsd.org X-Comment-To: "Anton A. Karpov" Message-ID: <20050424190912.GA6743@laptoxa.toxa.lan> Mail-Followup-To: freebsd-current@freebsd.org, freebsd-geom@freebsd.org, Pawel Jakub Dawidek References: <20050424022434.GA3384@laptoxa.toxa.lan> <20050424094343.GA837@darkness.comp.waw.pl> Mime-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Disposition: inline In-Reply-To: <20050424094343.GA837@darkness.comp.waw.pl> User-Agent: Outluck Express 1.5.6i for MS-DOS 6.22-SMP X-Mailer: See User-Agent above :) X-Operating-System: MS-DOS 6.22-CURRENT on Sony VAIO laptop X-PGP-Public-Key: http://toxahost.ru/gpg/pubkey.asc X-Useless-Header: Do Androids Dream of Electric Sheep? cc: Pawel Jakub Dawidek Subject: Re: gmirror: Not all disks connected X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 19:09:17 -0000 On Sun, Apr 24, 2005 at 11:43:43AM +0200, Pawel Jakub Dawidek wrote: > Please read description of 'forget' subcommand. Well, FYI, the problem *I THINK* was in the fact disks are not equal in size (ad1 is 3 sectors bigger): ad0: 38203MB [77619/16/63] at ata0-master UDMA100 ad1: 38204MB [77622/16/63] at ata0-slave UDMA100 In this case, fresh system was installed onto ad0, gm0 was created on ad1, when synchronise with added ad0 (as described in your step-by-step guide). The first synchronisation process (after addinbg ad0 into gm0) *was ok*, but after reboot, *I THINK*, gmirror discovered ad0 is less size than ad1 and refused to include it into gm0. Is this a bug or feature? The problem was solved by forget'ing gm0, creating gm1 on ad0, whet rebooting and adding ad1 (wthic is bigger than ad0) into it. Now everything seems to work ok... Thanks for the gmirror. From owner-freebsd-geom@FreeBSD.ORG Sun Apr 24 19:30:21 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 66E0516A4D1; Sun, 24 Apr 2005 19:30:21 +0000 (GMT) Received: from darkness.comp.waw.pl (darkness.comp.waw.pl [195.117.238.136]) by mx1.FreeBSD.org (Postfix) with ESMTP id A2A9243D39; Sun, 24 Apr 2005 19:30:20 +0000 (GMT) (envelope-from pjd@darkness.comp.waw.pl) Received: by darkness.comp.waw.pl (Postfix, from userid 1009) id BA935AC976; Sun, 24 Apr 2005 21:30:18 +0200 (CEST) Date: Sun, 24 Apr 2005 21:30:18 +0200 From: Pawel Jakub Dawidek To: freebsd-current@freebsd.org, freebsd-geom@freebsd.org Message-ID: <20050424193018.GH837@darkness.comp.waw.pl> References: <20050424022434.GA3384@laptoxa.toxa.lan> <20050424094343.GA837@darkness.comp.waw.pl> <20050424190912.GA6743@laptoxa.toxa.lan> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="XmZLSFikexiR11mx" Content-Disposition: inline In-Reply-To: <20050424190912.GA6743@laptoxa.toxa.lan> User-Agent: Mutt/1.4.2i X-PGP-Key-URL: http://people.freebsd.org/~pjd/pjd.asc X-OS: FreeBSD 5.2.1-RC2 i386 Subject: Re: gmirror: Not all disks connected X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Apr 2005 19:30:21 -0000 --XmZLSFikexiR11mx Content-Type: text/plain; charset=iso-8859-2 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sun, Apr 24, 2005 at 11:09:12PM +0400, Toxa wrote: +> On Sun, Apr 24, 2005 at 11:43:43AM +0200, Pawel Jakub Dawidek wrote: +>=20 +> > Please read description of 'forget' subcommand. +>=20 +> Well, FYI, the problem *I THINK* was in the fact disks are not equal in +> size (ad1 is 3 sectors bigger): +>=20 +> ad0: 38203MB [77619/16/63] at ata0-master UDM= A100 +> ad1: 38204MB [77622/16/63] at ata0-slave UDMA= 100 +>=20 +> In this case, fresh system was installed onto ad0, gm0 was created on ad= 1,=20 +> when synchronise with added ad0 (as described in your step-by-step guide= ). +> The first synchronisation process (after addinbg ad0 into gm0) *was ok*, +> but after reboot, *I THINK*, gmirror discovered ad0 is less size than +> ad1 and refused to include it into gm0. Is this a bug or feature? +>=20 +> The problem was solved by forget'ing gm0, creating gm1 on ad0, whet rebo= oting and adding +> ad1 (wthic is bigger than ad0) into it. Now everything seems to work +> ok... If you created gmirror giving first smaller disk and then connecting bigger one it should be ok. If disk which you want to insert is smaller than already existing components, gmirror should return an error, so this is probably not the case. PS. This step-by-step guide is not mine - Ralf S. Engelschall is its author. --=20 Pawel Jakub Dawidek http://www.wheel.pl pjd@FreeBSD.org http://www.FreeBSD.org FreeBSD committer Am I Evil? Yes, I Am! --XmZLSFikexiR11mx Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFCa/PKForvXbEpPzQRAtqcAKDhhu9thufeS1tvdrgjgy97AhrqhgCfamS8 J/91pXXis4fb8xm52IbB5m4= =67XO -----END PGP SIGNATURE----- --XmZLSFikexiR11mx-- From owner-freebsd-geom@FreeBSD.ORG Mon Apr 25 02:47:23 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3345E16A4CE for ; Mon, 25 Apr 2005 02:47:23 +0000 (GMT) Received: from zproxy.gmail.com (zproxy.gmail.com [64.233.162.206]) by mx1.FreeBSD.org (Postfix) with ESMTP id 7BF1E43D53 for ; Mon, 25 Apr 2005 02:47:22 +0000 (GMT) (envelope-from jclaeson@gmail.com) Received: by zproxy.gmail.com with SMTP id 9so894252nzo for ; Sun, 24 Apr 2005 19:47:22 -0700 (PDT) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:reply-to:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references; b=E9jBbw6obBRRsXmOZLs/SQ4uiBEdR96jM9V2O8AKpwxTZqqBmZUJccqViLUYuoCTPOwFs34JIX0XVJegue88usRQeLpAx9mnLxIQn57jTmJ10rFQrrHxcwQ9TdP4FOIBXK5p8fNIlau9FHVfD1r3r6OCYcjBEE/SIXdp51slFJE= Received: by 10.36.91.2 with SMTP id o2mr363454nzb; Sun, 24 Apr 2005 19:47:22 -0700 (PDT) Received: by 10.36.103.4 with HTTP; Sun, 24 Apr 2005 19:47:21 -0700 (PDT) Message-ID: <16a400ac050424194712d4eca0@mail.gmail.com> Date: Sun, 24 Apr 2005 22:47:21 -0400 From: Jeremy Claeson To: freebsd-geom@freebsd.org In-Reply-To: <16a400ac05041317272eb1ea01@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Content-Disposition: inline References: <4250A400.50200@videotron.ca> <4250F778.3040903@zvala.cz> <4251F631.3050707@videotron.ca> <20050404213800.A29631@denninger.net> <16a400ac05041317272eb1ea01@mail.gmail.com> Subject: Re: mirroring two data disks (no system files on them) X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list Reply-To: Jeremy Claeson List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Apr 2005 02:47:23 -0000 I'm still confused. I've seen some great emails on some things that seem pretty complicated to me, but I've yet to hear from anyone who can walk me through what I'm doing wrong. Maybe I'm doing something too complicated, can someone let me know that at least? Thanks! Please help and thanks in advance! On 4/13/05, Jeremy Claeson wrote: > I'm trying to do the same thing but I've not had any luck still. I'm > messing up somewhere in preparing the drives. I've got two 250GB > drives (ad4 and ad6). I'd like to mirror them as data disks. >=20 > Here's what I'm doing: > $ /stand/sysinstall > (fdisk and label ad4 as /usr/sambashares/data and writing that out to > disk. this works and I can access the disk) > $ /stand/sysinstall > (fdisk ad6 and write that out to disk) > $ gmirror label -v -n -b round-robin gm0s1 /dev/ad3s1 > $ gmirror load > (Here everything looks ok but I get a warning: > "WARNING: Expected rawoffset 0, found 63" > "GEOM_MIRROR: Cannot add disk mirror/gm0s1c to gm0s1 (error=3D17).") > $ gmirror configure -a gm0s1 > (at this point I get an error that says "Bad address" and then nothing > else works.) >=20 > [root@epiphyte jclaeson]# ls /dev/ad4* /dev/ad6* > /dev/ad4 /dev/ad4s1 /dev/ad4s1c /dev/ad4s1d /dev/ad6 >=20 > [root@epiphyte jclaeson]# ll /dev/mirror/ > total 1 > dr-xr-xr-x 2 root wheel 512 Apr 13 20:19 ./ > dr-xr-xr-x 5 root wheel 512 Apr 13 16:13 ../ > crw-r----- 1 root operator 4, 30 Apr 13 20:13 gm0 > crw-r----- 1 root operator 4, 31 Apr 13 20:13 gm0s1 > crw-r----- 1 root operator 4, 31 Apr 13 20:13 gm0s1 > crw-r----- 1 root operator 4, 33 Apr 13 20:13 gm0s1c > crw-r----- 1 root operator 4, 33 Apr 13 20:13 gm0s1c >=20 > [root@epiphyte jclaeson]# bsdlabel ad4s1 > # /dev/ad4s1: > 8 partitions: > # size offset fstype [fsize bsize bps/cpg] > c: 488392002 0 unused 0 0 # "raw" part, > don't edit > d: 488392002 0 4.2BSD 2048 16384 28552 >=20 > [root@epiphyte jclaeson]# bsdlabel ad6s1 > bsdlabel: /dev/ad6s1: No such file or directory > [root@epiphyte jclaeson]# bsdlabel ad6 > bsdlabel: /dev/ad6: no valid label found > [root@epiphyte jclaeson]# >=20 > So, I'm sure I'm preppring the disks wrong, I'm just not sure what I'm > doing wrong. >=20 > Thanks in advance for your help and advice! >=20 > ~jeremy >=20 >=20 > On 4/4/05, Karl Denninger wrote: > > On Mon, Apr 04, 2005 at 10:21:37PM -0400, Christian Caron wrote: > > > Tomas Zvala a =E9crit : > > > > did /dev/ad3s1 contain filesystem already? did you do newfs > > > > /dev/mirror/gm0s1 ? what does it mean "dies"? what does it say? > > > > > > Ok... man newfs shed some light on how to prepare a new disk... > > > > > > newfs -U /dev/mirror/gm0s1 > > > mkdir /mirror > > > mount /dev/mirror/gm0s1 /mirror > > > > > > did the trick. It boots ok and it seems to work (gmirror list does > > > return something, although it says "DIRTY" which doesn't sound > > > excellent...). > > > > > > I'll read the manual. > > > > > > Thanks! > > > > > > Christian > > > > "Dirty" is normal. It means the volume is mounted and has data on it, = and > > is not "clean", just as a mounted disk is "dirty" while it is actually > > mounted. > > > > -- > > -- > > Karl Denninger (karl@denninger.net) Internet Consultant & Kids Rights A= ctivist > > http://www.denninger.net My home on the net - links to everythin= g I do! > > http://scubaforum.org Your UNCENSORED place to talk about DIV= ING! > > http://www.spamcuda.net SPAM FREE mailboxes - FREE FOR A LIMITE= D TIME! > > http://genesis3.blogspot.com Musings Of A Sentient Mind > > > > > > _______________________________________________ > > freebsd-geom@freebsd.org mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-geom > > To unsubscribe, send any mail to "freebsd-geom-unsubscribe@freebsd.org" > > > From owner-freebsd-geom@FreeBSD.ORG Mon Apr 25 04:33:09 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3B6BB16A4CE for ; Mon, 25 Apr 2005 04:33:09 +0000 (GMT) Received: from gromit.dlib.vt.edu (gromit.dlib.vt.edu [128.173.49.29]) by mx1.FreeBSD.org (Postfix) with ESMTP id 9223443D41 for ; Mon, 25 Apr 2005 04:33:08 +0000 (GMT) (envelope-from paul@gromit.dlib.vt.edu) Received: from zappa.Chelsea-Ct.Org (pool-70-110-10-69.roa.east.verizon.net [70.110.10.69]) by gromit.dlib.vt.edu (8.13.3/8.13.3) with ESMTP id j3P4X6GA021290 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 25 Apr 2005 00:33:07 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) Received: from zappa.Chelsea-Ct.Org (localhost.Chelsea-Ct.Org [127.0.0.1]) by zappa.Chelsea-Ct.Org (8.13.3/8.13.3) with ESMTP id j3P4X0b6079390 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 25 Apr 2005 00:33:01 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) Received: (from paul@localhost) by zappa.Chelsea-Ct.Org (8.13.3/8.13.3/Submit) id j3P4Wxr2079389; Mon, 25 Apr 2005 00:33:00 -0400 (EDT) (envelope-from paul@gromit.dlib.vt.edu) X-Authentication-Warning: zappa.Chelsea-Ct.Org: paul set sender to paul@gromit.dlib.vt.edu using -f From: Paul Mather To: Jeremy Claeson In-Reply-To: <16a400ac050424194712d4eca0@mail.gmail.com> References: <4250A400.50200@videotron.ca> <4250F778.3040903@zvala.cz> <4251F631.3050707@videotron.ca> <20050404213800.A29631@denninger.net> <16a400ac05041317272eb1ea01@mail.gmail.com> <16a400ac050424194712d4eca0@mail.gmail.com> Content-Type: text/plain Content-Transfer-Encoding: 7bit Date: Mon, 25 Apr 2005 00:32:59 -0400 Message-Id: <1114403579.78883.80.camel@zappa.Chelsea-Ct.Org> Mime-Version: 1.0 X-Mailer: Evolution 2.2.2 FreeBSD GNOME Team Port cc: freebsd-geom@freebsd.org Subject: Re: mirroring two data disks (no system files on them) X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Apr 2005 04:33:09 -0000 On Sun, 2005-04-24 at 22:47 -0400, Jeremy Claeson wrote: > I'm still confused. I've seen some great emails on some things that > seem pretty complicated to me, but I've yet to hear from anyone who > can walk me through what I'm doing wrong. Maybe I'm doing something > too complicated, can someone let me know that at least? Thanks! Please > help and thanks in advance! > > On 4/13/05, Jeremy Claeson wrote: > > I'm trying to do the same thing but I've not had any luck still. I'm > > messing up somewhere in preparing the drives. I've got two 250GB > > drives (ad4 and ad6). I'd like to mirror them as data disks. > > > > Here's what I'm doing: > > $ /stand/sysinstall > > (fdisk and label ad4 as /usr/sambashares/data and writing that out to > > disk. this works and I can access the disk) > > $ /stand/sysinstall > > (fdisk ad6 and write that out to disk) > > $ gmirror label -v -n -b round-robin gm0s1 /dev/ad3s1 > > $ gmirror load > > (Here everything looks ok but I get a warning: > > "WARNING: Expected rawoffset 0, found 63" > > "GEOM_MIRROR: Cannot add disk mirror/gm0s1c to gm0s1 (error=17).") > > $ gmirror configure -a gm0s1 > > (at this point I get an error that says "Bad address" and then nothing > > else works.) > > > > [root@epiphyte jclaeson]# ls /dev/ad4* /dev/ad6* > > /dev/ad4 /dev/ad4s1 /dev/ad4s1c /dev/ad4s1d /dev/ad6 > > > > [root@epiphyte jclaeson]# ll /dev/mirror/ > > total 1 > > dr-xr-xr-x 2 root wheel 512 Apr 13 20:19 ./ > > dr-xr-xr-x 5 root wheel 512 Apr 13 16:13 ../ > > crw-r----- 1 root operator 4, 30 Apr 13 20:13 gm0 > > crw-r----- 1 root operator 4, 31 Apr 13 20:13 gm0s1 > > crw-r----- 1 root operator 4, 31 Apr 13 20:13 gm0s1 > > crw-r----- 1 root operator 4, 33 Apr 13 20:13 gm0s1c > > crw-r----- 1 root operator 4, 33 Apr 13 20:13 gm0s1c > > > > [root@epiphyte jclaeson]# bsdlabel ad4s1 > > # /dev/ad4s1: > > 8 partitions: > > # size offset fstype [fsize bsize bps/cpg] > > c: 488392002 0 unused 0 0 # "raw" part, > > don't edit > > d: 488392002 0 4.2BSD 2048 16384 28552 > > > > [root@epiphyte jclaeson]# bsdlabel ad6s1 > > bsdlabel: /dev/ad6s1: No such file or directory > > [root@epiphyte jclaeson]# bsdlabel ad6 > > bsdlabel: /dev/ad6: no valid label found > > [root@epiphyte jclaeson]# > > > > So, I'm sure I'm preppring the disks wrong, I'm just not sure what I'm > > doing wrong. In the above, you shouldn't be trying to bsdlabel the underlying providers after having created your mirror using them. Have you tried following the example in the gmirror man page? If you want to create a single filesystem covering the entire mirror composed of entire disks from which you don't need to boot, you can dispense with a partition table (slice) and even disklabel. I.e., you can skip fdisk and bsdlabel (or using sysinstall). In other words, to create a "data" mirror on ad4 and ad6, try something like: gmirror label -v -b round-robin data ad4 ad6 newfs -U /dev/mirror/data mount /dev/mirror/data /usr/sambashares/data (or wherever) You can add /dev/mirror/data to /etc/fstab, and add 'geom_mirror_load="YES"' to /boot/loader.conf to have the mirror mounted automatically at boot. (Or you can add "options GEOM_MIRROR" to your kernel config file instead of loading it as a kernel module via /boot/loader.conf.) If you have been experimenting with ad4 and ad6, you might want to dd /dev/zero over the beginning of the drive to get rid of any existing partition table or disklabel. You can also use "gmirror clear ad4 ad6" to clear any existing mirror metadata on there. If you want to have multiple file systems on the mirror, you might want to bsdlabel /dev/mirror/data between the "gmirror label" and newfs commands. Cheers, Paul. -- e-mail: paul@gromit.dlib.vt.edu "Without music to decorate it, time is just a bunch of boring production deadlines or dates by which bills must be paid." --- Frank Vincent Zappa From owner-freebsd-geom@FreeBSD.ORG Mon Apr 25 09:50:23 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id CCE7416A4CE; Mon, 25 Apr 2005 09:50:23 +0000 (GMT) Received: from mx.toxahost.ru (ns.toxahost.ru [62.89.204.62]) by mx1.FreeBSD.org (Postfix) with ESMTP id E30D243D4C; Mon, 25 Apr 2005 09:50:22 +0000 (GMT) (envelope-from toxa@toxahost.ru) Received: from localhost (laptoxa.toxa.lan [192.168.1.3]) by mx.toxahost.ru (Toxa) with ESMTP id D1D0691; Mon, 25 Apr 2005 13:50:20 +0400 (MSD) Date: Mon, 25 Apr 2005 13:50:16 +0400 From: Toxa To: freebsd-current@freebsd.org, freebsd-geom@freebsd.org X-Comment-To: "Anton A. Karpov" Message-ID: <20050425095016.GA10621@laptoxa.toxa.lan> Mail-Followup-To: freebsd-current@freebsd.org, freebsd-geom@freebsd.org References: <20050424022434.GA3384@laptoxa.toxa.lan> <20050424094343.GA837@darkness.comp.waw.pl> <20050424190912.GA6743@laptoxa.toxa.lan> <20050424193018.GH837@darkness.comp.waw.pl> Mime-Version: 1.0 Content-Type: text/plain; charset=koi8-r Content-Disposition: inline In-Reply-To: <20050424193018.GH837@darkness.comp.waw.pl> User-Agent: Outluck Express 1.5.6i for MS-DOS 6.22-SMP X-Mailer: See User-Agent above :) X-Operating-System: MS-DOS 6.22-CURRENT on Sony VAIO laptop X-PGP-Public-Key: http://toxahost.ru/gpg/pubkey.asc X-Useless-Header: Do Androids Dream of Electric Sheep? Subject: Re: gmirror: Not all disks connected X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Apr 2005 09:50:23 -0000 On Sun, Apr 24, 2005 at 09:30:18PM +0200, Pawel Jakub Dawidek wrote: > If you created gmirror giving first smaller disk and then connecting bigger > one it should be ok. If disk which you want to insert is smaller than > already existing components, gmirror should return an error, so this is > probably not the case. Well, as you can see, the disk I inserted was *a little* smaller, in fact, it should be equal size of the first one, and gmirror accepted it into gm0 *without any error*. But after final reboot it refused to attach it. Changing disk order (creating new gm1 on smaller disk, then attaching a bigger disk to gm1) solved this problem. So maybe this is a weird bug... From owner-freebsd-geom@FreeBSD.ORG Mon Apr 25 10:36:28 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id ADF2B16A4CE; Mon, 25 Apr 2005 10:36:28 +0000 (GMT) Received: from darkness.comp.waw.pl (darkness.comp.waw.pl [195.117.238.136]) by mx1.FreeBSD.org (Postfix) with ESMTP id A5E1F43D39; Mon, 25 Apr 2005 10:36:27 +0000 (GMT) (envelope-from pjd@darkness.comp.waw.pl) Received: by darkness.comp.waw.pl (Postfix, from userid 1009) id 900EBAC861; Mon, 25 Apr 2005 12:36:11 +0200 (CEST) Date: Mon, 25 Apr 2005 12:36:11 +0200 From: Pawel Jakub Dawidek To: freebsd-current@freebsd.org, freebsd-geom@freebsd.org Message-ID: <20050425103611.GJ837@darkness.comp.waw.pl> References: <20050424022434.GA3384@laptoxa.toxa.lan> <20050424094343.GA837@darkness.comp.waw.pl> <20050424190912.GA6743@laptoxa.toxa.lan> <20050424193018.GH837@darkness.comp.waw.pl> <20050425095016.GA10621@laptoxa.toxa.lan> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="9b4eu16IbjLAsbt2" Content-Disposition: inline In-Reply-To: <20050425095016.GA10621@laptoxa.toxa.lan> User-Agent: Mutt/1.4.2i X-PGP-Key-URL: http://people.freebsd.org/~pjd/pjd.asc X-OS: FreeBSD 5.2.1-RC2 i386 Subject: Re: gmirror: Not all disks connected X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 25 Apr 2005 10:36:28 -0000 --9b4eu16IbjLAsbt2 Content-Type: text/plain; charset=iso-8859-2 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Apr 25, 2005 at 01:50:16PM +0400, Toxa wrote: +> On Sun, Apr 24, 2005 at 09:30:18PM +0200, Pawel Jakub Dawidek wrote: +> > If you created gmirror giving first smaller disk and then connecting b= igger +> > one it should be ok. If disk which you want to insert is smaller than +> > already existing components, gmirror should return an error, so this is +> > probably not the case. +>=20 +>=20 +> Well, as you can see, the disk I inserted was *a little* smaller, in +> fact, it should be equal size of the first one, and gmirror accepted it +> into gm0 *without any error*. But after final reboot it refused to attac= h it. +> Changing disk order (creating new gm1 on smaller disk, then attaching=20 +> a bigger disk to gm1) solved this problem. So maybe this is a weird +> bug... I checked the code and there was a bug, but only when there was one sector difference. So if in your case was 3 sectors, it should tell you that your provider is too small: # mdconfig -a -t malloc -s 100 md0 # mdconfig -a -t malloc -s 103 md1 # gmirror label foo md1 # gmirror insert foo md0 Provider md0 too small. Exit 1 --=20 Pawel Jakub Dawidek http://www.wheel.pl pjd@FreeBSD.org http://www.FreeBSD.org FreeBSD committer Am I Evil? Yes, I Am! --9b4eu16IbjLAsbt2 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (FreeBSD) iD8DBQFCbMgbForvXbEpPzQRAqH8AJ9MUMVKiqT4zAgcNtjps7cjuvBEpQCeI7Tj dExq81y8xJriOSJ6CB1gmPA= =S4p5 -----END PGP SIGNATURE----- --9b4eu16IbjLAsbt2-- From owner-freebsd-geom@FreeBSD.ORG Tue Apr 26 13:01:42 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E41A116A4CE for ; Tue, 26 Apr 2005 13:01:42 +0000 (GMT) Received: from tomts10-srv.bellnexxia.net (tomts10.bellnexxia.net [209.226.175.54]) by mx1.FreeBSD.org (Postfix) with ESMTP id 30B3643D48 for ; Tue, 26 Apr 2005 13:01:42 +0000 (GMT) (envelope-from dlavigne6@sympatico.ca) Received: from [192.168.2.99] ([67.68.18.63]) by tomts10-srv.bellnexxia.net ESMTP <20050426130140.DPTP26102.tomts10-srv.bellnexxia.net@[192.168.2.99]> for ; Tue, 26 Apr 2005 09:01:40 -0400 Date: Tue, 26 Apr 2005 09:02:35 -0400 (EDT) From: Dru X-X-Sender: dlavigne6@dru.domain.org To: freebsd-geom@freebsd.org Message-ID: <20050426085330.T583@dru.domain.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Subject: offset 16 at bsdlabel step of gmirror config X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Apr 2005 13:01:43 -0000 I've found Ralf's tutorial on configuring gmirror and have read through the threads on this list. But I'm still unsure of the bsdlabel step when configuring gmirror. I know partition a is supposed to be offset 16 instead of 0. When I change it, I get this message: partitions a and b overlap! If I bump up b's offset by 16, that message goes away, but I still have this one: bsdlabel: partition c doesn't cover the whole unit! I get the same message if I bump up all the offsets (except c) by 16. Should I: 1. just bump a's offset up by 16 and ignore error messages? 2. bump a and b's offsets up by 16 and assume error message on c is due to sysinstall bug and can be ignored? 3. bump up all the offsets by 16 (except for c which shouldn't be touched) and ignore error message about c Dru From owner-freebsd-geom@FreeBSD.ORG Tue Apr 26 14:04:33 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 9524216A4CE for ; Tue, 26 Apr 2005 14:04:33 +0000 (GMT) Received: from koyukuk.teamcool.net (koyukuk.teamcool.net [209.161.34.19]) by mx1.FreeBSD.org (Postfix) with ESMTP id E70DB43D1D for ; Tue, 26 Apr 2005 14:04:32 +0000 (GMT) (envelope-from kgunders@teamcool.net) Received: from cochise.teamcool.net (unknown [192.168.1.57]) by koyukuk.teamcool.net (TeamCool Rocks) with ESMTP id E5B32F7E7 for ; Tue, 26 Apr 2005 08:04:31 -0600 (MDT) Date: Tue, 26 Apr 2005 08:04:31 -0600 From: Ken Gunderson To: freebsd-geom@freebsd.org Message-Id: <20050426080431.578642c8.kgunders@teamcool.net> In-Reply-To: <20050426085330.T583@dru.domain.org> References: <20050426085330.T583@dru.domain.org> Organization: Teamcool Networks X-Mailer: Sylpheed version 1.9.6 (GTK+ 2.6.4; i386-portbld-freebsd5.3) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: Re: offset 16 at bsdlabel step of gmirror config X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Apr 2005 14:04:33 -0000 On Tue, 26 Apr 2005 09:02:35 -0400 (EDT) Dru wrote: > > I've found Ralf's tutorial on configuring gmirror and have read through > the threads on this list. But I'm still unsure of the bsdlabel step > when configuring gmirror. > > I know partition a is supposed to be offset 16 instead of 0. When I change > it, I get this message: > > partitions a and b overlap! > > If I bump up b's offset by 16, that message goes away, but I still have > this one: > > bsdlabel: partition c doesn't cover the whole unit! > > I get the same message if I bump up all the offsets (except c) by 16. > > Should I: > > 1. just bump a's offset up by 16 and ignore error messages? > 2. bump a and b's offsets up by 16 and assume error message on c is due to > sysinstall bug and can be ignored? > 3. bump up all the offsets by 16 (except for c which shouldn't be touched) > and ignore error message about c > > Dru When you increase partition a's offset by 16 you ALSO need to subtract 16 from the size of partition a. Not doing so was what was causing partitiona a and b to overlap. That's the simplest way anyhow. Have fun. -- Best regards, Ken Gunderson GPG Key-- 9F5179FD "Freedom begins between the ears." -- Edward Abbey From owner-freebsd-geom@FreeBSD.ORG Tue Apr 26 18:58:07 2005 Return-Path: Delivered-To: freebsd-geom@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 2ED7116A4CE for ; Tue, 26 Apr 2005 18:58:07 +0000 (GMT) Received: from tomts22-srv.bellnexxia.net (tomts22.bellnexxia.net [209.226.175.184]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1E75F43D48 for ; Tue, 26 Apr 2005 18:58:06 +0000 (GMT) (envelope-from dlavigne6@sympatico.ca) Received: from [192.168.2.99] ([67.68.18.63]) by tomts22-srv.bellnexxia.net ESMTP <20050426185805.LMWU21470.tomts22-srv.bellnexxia.net@[192.168.2.99]> for ; Tue, 26 Apr 2005 14:58:05 -0400 Date: Tue, 26 Apr 2005 14:58:59 -0400 (EDT) From: Dru X-X-Sender: dlavigne6@dru.domain.org To: freebsd-geom@freebsd.org Message-ID: <20050426144746.T583@dru.domain.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Subject: cannot open consumer X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Apr 2005 18:58:07 -0000 Following Ralf's how-to. Successfully ran: gmirror label -v -n -h -b round-robin gm0s1 /dev/ad2s1 Get this message on gmirror load: GEOM_MIRROR: Device gm0s1 created (id=1024834944). GEOM_MIRROR: Device gm0s1: provider ad2s1 detected. GEOM_MIRROR: Device gm0s1: provider ad2s1 activated. GEOM_MIRROR: Device gm0s1: provider mirror/gm0s1 launched. GEOM_MIRROR: Device gm0 created (id=1640242348). GEOM_MIRROR: Cannot open consumer ad0 (error=1). GEOM_MIRROR: Cannot add disk ad0 to gm0 (error=1). GEOM_MIRROR: Device gm0 destroyed. I'm assuming that's not good. If I ignore that and carry on and bsdlabel /dev/mirror/gm0s1 so it looks exactly like /dev/ad0s1 (except a has offset of 16 and a's size is decreased by 16), I again run into problems when I go to insert the first drive into the mirror: gmirror insert gm0s1 /dev/ad0s1 Provider ad0s1 too small. Which is frustrating seeing the 2 drives are identical hardware, fdisk'd identically and labelled identically (except the offset 16 thing). Suggestions appreciated. Here's uname: FreeBSD some.host.org 5.4-RC3 FreeBSD 5.4-RC3 #0: Fri Apr 22 09:02:55 EDT 2005 root@some.host.org:/usr/obj/usr/src/sys/NEW i386 Dru