From owner-freebsd-geom@FreeBSD.ORG Sun Jan 14 00:31:11 2007 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 4CCCA16A417 for ; Sun, 14 Jan 2007 00:31:11 +0000 (UTC) (envelope-from gcubfg-freebsd-geom@m.gmane.org) Received: from ciao.gmane.org (main.gmane.org [80.91.229.2]) by mx1.freebsd.org (Postfix) with ESMTP id D461013C45B for ; Sun, 14 Jan 2007 00:31:10 +0000 (UTC) (envelope-from gcubfg-freebsd-geom@m.gmane.org) Received: from list by ciao.gmane.org with local (Exim 4.43) id 1H5tH5-0007YY-Gy for freebsd-geom@freebsd.org; Sun, 14 Jan 2007 01:31:03 +0100 Received: from p3ee2281c.dip0.t-ipconnect.de ([62.226.40.28]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sun, 14 Jan 2007 01:31:03 +0100 Received: from christian.baer by p3ee2281c.dip0.t-ipconnect.de with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sun, 14 Jan 2007 01:31:03 +0100 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-geom@freebsd.org From: Christian Baer Date: Sun, 14 Jan 2007 01:26:02 +0100 (CET) Organization: Convenimus Projekt Lines: 43 Message-ID: References: <20070113202018.GK90718@garage.freebsd.pl> X-Complaints-To: usenet@sea.gmane.org X-Gmane-NNTP-Posting-Host: p3ee2281c.dip0.t-ipconnect.de User-Agent: slrn/0.9.8.1 (FreeBSD) Sender: news Subject: Re: What does geli attach -a do? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 14 Jan 2007 00:31:11 -0000 On Sat, 13 Jan 2007 21:20:18 +0100 Pawel Jakub Dawidek wrote: > It'll tell you exact offset and size where corrupted data were detected. > It won't help you bring you data back, it's a security feature, not a > reliability feature, but can be used also to detect silent data > corruptions. Does that mean, it could probably tell me what file was affected >> Does it make sense to use this in combination with a mirror? > If you're afraid of silent data corruptions, then yes. When one half of > the mirror will be corrupted and geli will detect it, gmirror will read > the data from the other half. I'm not more afraid of those than of any other sort of breakdowns. :-) These are computers. They work but they also fail from time to time. I've lost quite a few HDs during the years that I've been working with computers now. Not all of the HDs went on friendly terms. :-) Basicly, the information is quite important but subject to constant changes. That is why I want to mirror the data, so I can have a pretty good chance of getting the changes back that were made between to backups. The question if this made sense in combination with a mirror went in another direction: Can gmirror detect which drive is holding the "right" data and retrieve it should a drive die? > Unfortunately authentication-only mode is not supported in geli at the > moment, so you have encryption/decription overhead. > If you don't care about this overhead, and don't care about security, > this is how you can create such configuration: Well, actually, I do care for security. Geli is already set up on the mirrors. I'm a little worried that this feature will however run the crap out of the CPUs. :-) This is a Sun U60 we are talking about. Not really what you'd call a CPU-monster. Copying a big file over the network onto an encrypted partition runs at about 5-6MB/s and uses 80-90% of both CPUs. There aren't many reserves to go around while doing that. Regards Chris From owner-freebsd-geom@FreeBSD.ORG Sun Jan 14 01:12:53 2007 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 7900716A412 for ; Sun, 14 Jan 2007 01:12:53 +0000 (UTC) (envelope-from cyberleo@cyberleo.net) Received: from pizzabox.cyberleo.net (alpha.cyberleo.net [198.145.45.10]) by mx1.freebsd.org (Postfix) with ESMTP id 39AD113C45B for ; Sun, 14 Jan 2007 01:12:53 +0000 (UTC) (envelope-from cyberleo@cyberleo.net) Received: (qmail 68332 invoked from network); 14 Jan 2007 01:12:52 -0000 Received: from adsl-69-212-1-127.dsl.chcgil.ameritech.net (HELO ?172.16.44.14?) (cyberleo@cyberleo.net@69.212.1.127) by alpha.cyberleo.net with ESMTPA; 14 Jan 2007 01:12:52 -0000 Message-ID: <45A9836B.5030802@cyberleo.net> Date: Sat, 13 Jan 2007 19:12:11 -0600 From: CyberLeo Kitsana User-Agent: Thunderbird 1.5 (X11/20051201) MIME-Version: 1.0 To: "R. B. Riddick" References: <376348.20105.qm@web30301.mail.mud.yahoo.com> In-Reply-To: <376348.20105.qm@web30301.mail.mud.yahoo.com> Content-Type: text/plain; charset=iso-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: FreeBSD Geom Subject: Re: geom_raid5 livelock? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 14 Jan 2007 01:12:53 -0000 R. B. Riddick wrote: > What happened to our little debug message project (lets call it "Preparation > D")? I mean: The debug message in case of a parity mismatch during a verify > read... I haven't forgotten. I was just trying to figure out how to build and use the new module without having to reboot the box. Guess there's no safe way. http://home.cyberleo.net/cyberleo/workspace/Zip/graid5-all3.log That's the latest log, with the debug lines inserted. Thanks! -- Fuzzy love, -CyberLeo Technical Administrator CyberLeo.Net Webhosting http://www.CyberLeo.Net Furry Peace! - http://www.fur.com/peace/ From owner-freebsd-geom@FreeBSD.ORG Sun Jan 14 01:26:21 2007 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 4E37F16A40F for ; Sun, 14 Jan 2007 01:26:21 +0000 (UTC) (envelope-from arne_woerner@yahoo.com) Received: from web30305.mail.mud.yahoo.com (web30305.mail.mud.yahoo.com [209.191.69.67]) by mx1.freebsd.org (Postfix) with SMTP id 155D113C428 for ; Sun, 14 Jan 2007 01:26:20 +0000 (UTC) (envelope-from arne_woerner@yahoo.com) Received: (qmail 95887 invoked by uid 60001); 14 Jan 2007 01:26:20 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=Message-ID:X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding; b=4TmBN3faWsnb+O8ENnN++LB7A2Q9ndU9vybYbUiMCwLRMfCjS3ZdNOUVULkCo2QiTK+9Opb4LM/emCW/6YpHl2LFxLd2q9wbfunzTVfImEiHV+IPNelS+CdjeY+UJtB739ZXLH2XJYhlW5EWgL6WnhLH+nRz7Cevv5LAwGGIbC4= ; Message-ID: <20070114012620.95885.qmail@web30305.mail.mud.yahoo.com> X-YMail-OSG: Rnhm.3cVM1kac56vzWb1a6Vv03UTTLN5P_355TH1F3fO7D7ocajGFyTQ2gMpKCZTxqydu9VWsiTedc2d7Cfu9x4wNmHsYpnLXta3UDu2yYdPG9dZcViGOjBoVy7vYdS8_Chrrw29.X7fTjA- Received: from [213.54.176.27] by web30305.mail.mud.yahoo.com via HTTP; Sat, 13 Jan 2007 17:26:20 PST Date: Sat, 13 Jan 2007 17:26:20 -0800 (PST) From: "R. B. Riddick" To: CyberLeo Kitsana In-Reply-To: <45A9836B.5030802@cyberleo.net> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Cc: FreeBSD Geom Subject: Re: geom_raid5 livelock? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 14 Jan 2007 01:26:21 -0000 Thank you for the tests... :-) --- CyberLeo Kitsana wrote: > I haven't forgotten. I was just trying to figure out how to build and > use the new module without having to reboot the box. Guess there's no > safe way. > Yup... I always remove all disks, when I am testing (then I can unload the module)... But then u had to do a rebuild, which is not so good... I do not know, why GEOM taste-s the disks again, after the device has been stopped... I havent found out the trick yet... > http://home.cyberleo.net/cyberleo/workspace/Zip/graid5-all3.log > That's the latest log, with the debug lines inserted. > OK - So we have a false parity at 264438038528. Now we could 1. (hard way) Look at the data (16KB) at that offset of all disks and see, where exactly the parity mismatches ("dd if=/dev/adXs2 bs=16k iseek=16140017 of=/tmp/adXs2.test..."; then we could XOR them all together and see, where we have non-zero bits) or 2. (easy way) Do a re-sync (graid5 configure -R raid5) and look at kern.geom.raid5.veri_w (the number of bad parity blocks). The re-sync seems to be necessary, because it is very likely, that ur graid5 device has bad parity blocks... It is difficult to say, how that happened (could be an unintentional "graid5 configure -R" or a bug in graid5 or...)... If you can reproduce it (maybe with a smaller device?), I would be glad, if you could let me know, how u did it... -Arne ____________________________________________________________________________________ Yahoo! Music Unlimited Access over 1 million songs. http://music.yahoo.com/unlimited From owner-freebsd-geom@FreeBSD.ORG Sun Jan 14 03:04:00 2007 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 1EE4716A407 for ; Sun, 14 Jan 2007 03:04:00 +0000 (UTC) (envelope-from cyberleo@cyberleo.net) Received: from pizzabox.cyberleo.net (alpha.cyberleo.net [198.145.45.10]) by mx1.freebsd.org (Postfix) with ESMTP id D439E13C441 for ; Sun, 14 Jan 2007 03:03:59 +0000 (UTC) (envelope-from cyberleo@cyberleo.net) Received: (qmail 70165 invoked from network); 14 Jan 2007 03:03:59 -0000 Received: from adsl-69-212-1-127.dsl.chcgil.ameritech.net (HELO ?172.16.44.14?) (cyberleo@cyberleo.net@69.212.1.127) by alpha.cyberleo.net with ESMTPA; 14 Jan 2007 03:03:59 -0000 Message-ID: <45A99D5A.3060400@cyberleo.net> Date: Sat, 13 Jan 2007 21:02:50 -0600 From: CyberLeo Kitsana User-Agent: Thunderbird 1.5 (X11/20051201) MIME-Version: 1.0 To: "R. B. Riddick" References: <20070114012620.95885.qmail@web30305.mail.mud.yahoo.com> In-Reply-To: <20070114012620.95885.qmail@web30305.mail.mud.yahoo.com> Content-Type: text/plain; charset=iso-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: FreeBSD Geom Subject: Re: geom_raid5 livelock? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 14 Jan 2007 03:04:00 -0000 R. B. Riddick wrote: > Thank you for the tests... :-) You're welcome. Always glad to help squash bugs. > 1. (hard way) > Look at the data (16KB) at that offset of all disks and see, where exactly the > parity mismatches ("dd if=/dev/adXs2 bs=16k iseek=16140017 > of=/tmp/adXs2.test..."; then we could XOR them all together and see, where we > have non-zero bits) I XORed together all four 16k blocks captured during a livelock, and it came out to a file full of nul bytes. Given what you'd said, I believe that means the parity is correct for that stripe. -- Fuzzy love, -CyberLeo Technical Administrator CyberLeo.Net Webhosting http://www.CyberLeo.Net Furry Peace! - http://www.fur.com/peace/ From owner-freebsd-geom@FreeBSD.ORG Sun Jan 14 14:32:10 2007 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 1FBAB16A5BD for ; Sun, 14 Jan 2007 14:32:10 +0000 (UTC) (envelope-from r.c.ladan@gmail.com) Received: from nf-out-0910.google.com (nf-out-0910.google.com [64.233.182.189]) by mx1.freebsd.org (Postfix) with ESMTP id ADE8913C45A for ; Sun, 14 Jan 2007 14:32:09 +0000 (UTC) (envelope-from r.c.ladan@gmail.com) Received: by nf-out-0910.google.com with SMTP id k27so1773084nfc for ; Sun, 14 Jan 2007 06:32:08 -0800 (PST) DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:user-agent:mime-version:to:cc:subject:references:in-reply-to:x-enigmail-version:content-type:content-transfer-encoding; b=mKB/HhnqETgdQd75IPaa29ZFRvXYCwUw3XrKIKVmjaHsNvZdSeGuX3oyHGxGgu8pJfzDn4Jz8zoAnKoU7Yiu4LQijsKjrngJfkB/Nc45aueGOFvkOL7iqNlINMcSr8jKFTpTHaXH0gJeW0XWZ50IYisoHnCEKqFtmIvLW7SKoYM= Received: by 10.49.13.19 with SMTP id q19mr2952172nfi.1168785128233; Sun, 14 Jan 2007 06:32:08 -0800 (PST) Received: from ?192.168.123.201? ( [195.241.221.201]) by mx.google.com with ESMTP id k23sm17768814nfc.2007.01.14.06.32.06; Sun, 14 Jan 2007 06:32:07 -0800 (PST) Message-ID: <45AA3EDE.6070806@gmail.com> Date: Sun, 14 Jan 2007 15:31:58 +0100 From: Rene Ladan User-Agent: Thunderbird 1.5.0.9 (X11/20061224) MIME-Version: 1.0 To: Poul-Henning Kamp References: <14692.1168184789@critter.freebsd.dk> In-Reply-To: <14692.1168184789@critter.freebsd.dk> X-Enigmail-Version: 0.94.0.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: freebsd-geom@freebsd.org Subject: Re: pp->mediasize == 0 when reattaching physical medium ? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 14 Jan 2007 14:32:10 -0000 Poul-Henning Kamp schreef: > In message <45A103F4.5040108@gmail.com>, Rene Ladan writes: > >>> 1) If you kldload the class and plug in a physical medium afterwards, >>> then pp->mediasize == 0 so the taste function fails. > > That's a USB bug. > Well, I put the raw image of the medium on a CD to see if that works. Results: * if the CD is loaded during boot and the module is loaded after boot, the CD gets recognized by GEOM and the xbox360 class creates the two slices; * if the CD is loaded after boot and after the module is loaded, nothing happens (the drive spins up a few times, but the taste function is not called). So maybe the bug is somewhere in the CAM layer? Regards, Rene -- GPG fingerprint = E738 5471 D185 7013 0EE0 4FC8 3C1D 6F83 12E1 84F6 (subkeys.pgp.net) "It won't fit on the line." -- me, 2001 From owner-freebsd-geom@FreeBSD.ORG Sun Jan 14 23:12:47 2007 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 605F816A412 for ; Sun, 14 Jan 2007 23:12:47 +0000 (UTC) (envelope-from arne_woerner@yahoo.com) Received: from web30301.mail.mud.yahoo.com (web30301.mail.mud.yahoo.com [209.191.69.63]) by mx1.freebsd.org (Postfix) with SMTP id 0FA5F13C44C for ; Sun, 14 Jan 2007 23:12:46 +0000 (UTC) (envelope-from arne_woerner@yahoo.com) Received: (qmail 61958 invoked by uid 60001); 14 Jan 2007 23:12:46 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:Date:From:Subject:To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=HnsjxcqIfZ9h/SKGjwMhgZtZmrMhWfBQcn/FY9BAMpuS+pgpkfm3cEg+Nvc4Nk776bTQXTLaCWCPArf877P5XzGNJPECIwBMCHIu+WIUZckqE/8gDizbA9QQ0oJh/aMoXuEjVSwulLcT9ZX6LFUEx/Q7mgXAOb6Eljescp2MIaY=; X-YMail-OSG: 3ixLCToVM1nTxJo498CXvk2mZXrebqhoPuHxuuQGX0oh4TxtfbJMdY.mbU5Yu4LFg.YvAZurhm04RBs4Qv6500ZDh.ip7TNus__C0xVgdUGofuVk61rsmhCV51Ui8O96zEBgOntNnuuDHA-- Received: from [85.212.14.157] by web30301.mail.mud.yahoo.com via HTTP; Sun, 14 Jan 2007 15:12:45 PST Date: Sun, 14 Jan 2007 15:12:45 -0800 (PST) From: "R. B. Riddick" To: CyberLeo Kitsana , FreeBSD Geom MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <887701.59859.qm@web30301.mail.mud.yahoo.com> Cc: Subject: Re: geom_raid5 livelock? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 14 Jan 2007 23:12:47 -0000 It looks like, always the same consumer returns false data again and again in this strange situation, although at the same time a dd to the same consumer at the same offset returns data, that fits to the parity block. Does somebody here have an idea, why GEOM does that? Could it be, that graid5 ruined somehow memory management? Could it be, that GEOM is disturbed by simultaneous request? -Arne ____________________________________________________________________________________ It's here! Your new message! Get new email alerts with the free Yahoo! Toolbar. http://tools.search.yahoo.com/toolbar/features/mail/ From owner-freebsd-geom@FreeBSD.ORG Mon Jan 15 11:08:10 2007 Return-Path: X-Original-To: freebsd-geom@FreeBSD.org Delivered-To: freebsd-geom@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 9DD3916A492 for ; Mon, 15 Jan 2007 11:08:10 +0000 (UTC) (envelope-from owner-bugmaster@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [69.147.83.40]) by mx1.freebsd.org (Postfix) with ESMTP id 8AD8313C441 for ; Mon, 15 Jan 2007 11:08:10 +0000 (UTC) (envelope-from owner-bugmaster@FreeBSD.org) Received: from freefall.freebsd.org (linimon@localhost [127.0.0.1]) by freefall.freebsd.org (8.13.4/8.13.4) with ESMTP id l0FB8A8J031689 for ; Mon, 15 Jan 2007 11:08:10 GMT (envelope-from owner-bugmaster@FreeBSD.org) Received: (from linimon@localhost) by freefall.freebsd.org (8.13.4/8.13.4/Submit) id l0FB895Z031685 for freebsd-geom@FreeBSD.org; Mon, 15 Jan 2007 11:08:09 GMT (envelope-from owner-bugmaster@FreeBSD.org) Date: Mon, 15 Jan 2007 11:08:09 GMT Message-Id: <200701151108.l0FB895Z031685@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: linimon set sender to owner-bugmaster@FreeBSD.org using -f From: FreeBSD bugmaster To: freebsd-geom@FreeBSD.org Cc: Subject: Current problem reports assigned to you X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 15 Jan 2007 11:08:10 -0000 Current FreeBSD problem reports Critical problems Serious problems S Tracker Resp. Description -------------------------------------------------------------------------------- o kern/73177 geom kldload geom_* causes panic due to memory exhaustion o kern/76538 geom [gbde] nfs-write on gbde partition stalls and continue o kern/83464 geom [geom] [patch] Unhandled malloc failures within libgeo o kern/84556 geom [geom] GBDE-encrypted swap causes panic at shutdown o kern/87544 geom [gbde] mmaping large files on a gbde filesystem deadlo o kern/89102 geom [geom_vfs] [panic] panic when forced unmount FS from u o bin/90093 geom fdisk(8) incapable of altering in-core geometry o kern/90582 geom [geom_mirror] [panic] Restore cause panic string (ffs_ o kern/98034 geom [geom] dereference of NULL pointer in acd_geom_detach o kern/104389 geom [geom] [patch] sys/geom/geom_dump.c doesn't encode XML 10 problems total. Non-critical problems S Tracker Resp. Description -------------------------------------------------------------------------------- o bin/78131 geom gbde "destroy" not working. o kern/79251 geom [2TB] newfs fails on 2.6TB gbde device o kern/94632 geom [geom] Kernel output resets input while GELI asks for f kern/105390 geom [geli] filesystem on a md backed by sparse file with s o kern/107707 geom [geom] [patch] add new class geom_xbox360 to slice up 5 problems total. From owner-freebsd-geom@FreeBSD.ORG Fri Jan 19 15:58:33 2007 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 774DD16A402 for ; Fri, 19 Jan 2007 15:58:33 +0000 (UTC) (envelope-from bashr@comcast.net) Received: from sccrmhc13.comcast.net (sccrmhc13.comcast.net [204.127.200.83]) by mx1.freebsd.org (Postfix) with ESMTP id 44B7413C46A for ; Fri, 19 Jan 2007 15:58:33 +0000 (UTC) (envelope-from bashr@comcast.net) Received: from [149.28.227.65] (unknown[149.28.227.65]) by comcast.net (sccrmhc13) with ESMTP id <2007011915583201300l4gahe>; Fri, 19 Jan 2007 15:58:32 +0000 Message-ID: <45B0EAA7.7020609@comcast.net> Date: Fri, 19 Jan 2007 10:58:31 -0500 From: bashr User-Agent: Thunderbird 1.5.0.9 (X11/20061228) MIME-Version: 1.0 CC: freebsd-geom@freebsd.org References: <20070113004728.GQ2616@obiwan.tataz.chchile.org> <45A85F40.7050703@centtech.com> <45A8B722.7020302@fluffles.net> In-Reply-To: <45A8B722.7020302@fluffles.net> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: What is gstripe ? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: bashr@comcast.net List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 19 Jan 2007 15:58:33 -0000 Fluffles wrote: > concurrency. It's very important to select a high stripe size though; if > you lower the stripesize to say 16KB the parallellisation effect will be > almost gone: > How do you decide what stripe size to use? Are there handy rules-of-thumb? I must have used the default stripe size when setting up gstripe here, 4096. There are two gstripes here, which are then gmirrored. Are there similar effects of gmirror sector size? From owner-freebsd-geom@FreeBSD.ORG Sat Jan 20 00:59:34 2007 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 83CC516A400 for ; Sat, 20 Jan 2007 00:59:34 +0000 (UTC) (envelope-from etc@fluffles.net) Received: from auriate.fluffles.net (cust.95.160.adsl.cistron.nl [195.64.95.160]) by mx1.freebsd.org (Postfix) with ESMTP id 3890613C459 for ; Sat, 20 Jan 2007 00:59:34 +0000 (UTC) (envelope-from etc@fluffles.net) Received: from destiny ([10.0.0.21]) by auriate.fluffles.net with esmtpa (Exim 4.63 (FreeBSD)) (envelope-from ) id 1H84Zr-000Mih-Np; Sat, 20 Jan 2007 01:59:27 +0100 Message-ID: <45B169AB.3070103@fluffles.net> Date: Sat, 20 Jan 2007 02:00:27 +0100 From: Fluffles User-Agent: Thunderbird 1.5.0.8 (X11/20061114) MIME-Version: 1.0 To: bashr@comcast.net References: <20070113004728.GQ2616@obiwan.tataz.chchile.org> <45A85F40.7050703@centtech.com> <45A8B722.7020302@fluffles.net> <45B0EAA7.7020609@comcast.net> In-Reply-To: <45B0EAA7.7020609@comcast.net> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: freebsd-geom@freebsd.org Subject: Re: What is gstripe ? X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 20 Jan 2007 00:59:34 -0000 bashr wrote: > Fluffles wrote: >> concurrency. It's very important to select a high stripe size though; if >> you lower the stripesize to say 16KB the parallellisation effect will be >> almost gone: >> > > How do you decide what stripe size to use? Are there handy > rules-of-thumb? I must have used the default stripe size when setting > up gstripe here, 4096. There are two gstripes here, which are then > gmirrored. Are there similar effects of gmirror sector size? 4KB is an extremely low stripesize and will mostly likely degrade performance, howeveri suspect the effect would be less on RAID1-volumes since the RAID-subsystem does not have to switch devices when stripe blocks overlap. So RAID1 might be an exception, though also for RAID1 i think a stripesize of 128KB is best suited. Overall i sugest using a stripesize of 128KB of bigger. This way you can be sure that a single I/O request (maximum of 128KB on FreeBSD; MAXPHYS) will 'fit' into one stripeblock and thus can be handled by one disk in the RAID array. If you use 64KB stripesize and you read 65KB or 100KB, two physical disks must be used to handle the request; this will degrade performance. Misalignment, often caused by using default partitioning, can also degrade performance. To counteract this, use manual disklabeling with the same offset (or multiple) as the stripesize, use Dangerously Dedicated mode, or simply select a stripesize of 256KB or even bigger. A lower stripesize can in reality only be useful for combining software parity RAID. In all my benchmarks, a large stripesize was always faster than a small stripesize. So the bottom line is probably: use 128KB or larger. The sector size is the minimum transfersize; you should leave it at 512 bytes; except for special hardware or to speed up sector-based encryption such as GELI. - Veronica From owner-freebsd-geom@FreeBSD.ORG Sat Jan 20 08:31:22 2007 Return-Path: X-Original-To: freebsd-geom@freebsd.org Delivered-To: freebsd-geom@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 17A4216A400 for ; Sat, 20 Jan 2007 08:31:22 +0000 (UTC) (envelope-from xride@x12.dk) Received: from swip.net (mailfe12.tele2.dk [212.247.155.99]) by mx1.freebsd.org (Postfix) with ESMTP id AF26313C45E for ; Sat, 20 Jan 2007 08:31:21 +0000 (UTC) (envelope-from xride@x12.dk) X-Cloudmark-Score: 0.000000 [] Received: from x12.dk (account mu12272@get2net.dk [83.72.97.231] verified) by mailfe12.swip.net (CommuniGate Pro SMTP 5.0.12) with ESMTPA id 218220494 for freebsd-geom@freebsd.org; Sat, 20 Jan 2007 08:31:18 +0100 Received: by x12.dk (Postfix, from userid 666) id CFA9A50839; Sat, 20 Jan 2007 08:31:17 +0100 (CET) Date: Sat, 20 Jan 2007 08:31:17 +0100 From: Soeren Straarup To: freebsd-geom@freebsd.org Message-ID: <20070120073117.GB60202@x12.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.2.2i Subject: A beginner project X-BeenThere: freebsd-geom@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: GEOM-specific discussions and implementations List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 20 Jan 2007 08:31:22 -0000 What would be a beignner project within geom? Something is acutally going to be used by atleat some. I have briefly looked at: http://www.freebsd.org/projects/ideas/#p-psched But are there already someone out there that is working on it? /Soeren -- Soeren Straarup | aka OZ2DAK aka Xride FreeBSD committer | FreeBSD since 2.2.6-R If a program is not working right, then send a patch