From owner-freebsd-fs@FreeBSD.ORG Mon Aug 13 11:08:17 2007 Return-Path: Delivered-To: freebsd-fs@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id DDF3116A41B for ; Mon, 13 Aug 2007 11:08:17 +0000 (UTC) (envelope-from owner-bugmaster@FreeBSD.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:4f8:fff6::28]) by mx1.freebsd.org (Postfix) with ESMTP id CE3D713C4A8 for ; Mon, 13 Aug 2007 11:08:17 +0000 (UTC) (envelope-from owner-bugmaster@FreeBSD.org) Received: from freefall.freebsd.org (gnats@localhost [127.0.0.1]) by freefall.freebsd.org (8.14.1/8.14.1) with ESMTP id l7DB8HHd047664 for ; Mon, 13 Aug 2007 11:08:17 GMT (envelope-from owner-bugmaster@FreeBSD.org) Received: (from gnats@localhost) by freefall.freebsd.org (8.14.1/8.14.1/Submit) id l7DB8Gxx047660 for freebsd-fs@FreeBSD.org; Mon, 13 Aug 2007 11:08:16 GMT (envelope-from owner-bugmaster@FreeBSD.org) Date: Mon, 13 Aug 2007 11:08:16 GMT Message-Id: <200708131108.l7DB8Gxx047660@freefall.freebsd.org> X-Authentication-Warning: freefall.freebsd.org: gnats set sender to owner-bugmaster@FreeBSD.org using -f From: FreeBSD bugmaster To: freebsd-fs@FreeBSD.org Cc: Subject: Current problem reports assigned to you X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Aug 2007 11:08:18 -0000 Current FreeBSD problem reports Critical problems Serious problems S Tracker Resp. Description -------------------------------------------------------------------------------- o kern/112658 fs [smbfs] [patch] smbfs and caching problems (resolves b o kern/114676 fs [ufs] snapshot creation panics: snapacct_ufs2: bad blo o kern/114856 fs [ntfs] [patch] Bug in NTFS allows bogus file modes. o bin/115165 fs [PATCH] amd(8): add functionality of mount_nfs' -L -a 4 problems total. Non-critical problems S Tracker Resp. Description -------------------------------------------------------------------------------- o kern/114847 fs [ntfs] [patch] dirmask support for NTFS ala MSDOSFS 1 problem total. From owner-freebsd-fs@FreeBSD.ORG Tue Aug 14 07:00:31 2007 Return-Path: Delivered-To: fs@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 1A98A16A469; Tue, 14 Aug 2007 07:00:31 +0000 (UTC) (envelope-from brde@optusnet.com.au) Received: from mail14.syd.optusnet.com.au (mail14.syd.optusnet.com.au [211.29.132.195]) by mx1.freebsd.org (Postfix) with ESMTP id 344FC13C4A8; Tue, 14 Aug 2007 07:00:29 +0000 (UTC) (envelope-from brde@optusnet.com.au) Received: from besplex.bde.org (c220-239-235-248.carlnfd3.nsw.optusnet.com.au [220.239.235.248]) by mail14.syd.optusnet.com.au (8.13.1/8.13.1) with ESMTP id l7E70GHS006592 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Tue, 14 Aug 2007 17:00:27 +1000 Date: Tue, 14 Aug 2007 16:06:20 +1000 (EST) From: Bruce Evans X-X-Sender: bde@besplex.bde.org To: Kostik Belousov In-Reply-To: <20070810124153.GW2738@deviant.kiev.zoral.com.ua> Message-ID: <20070814154812.J24186@besplex.bde.org> References: <20070712084115.GA2200@deviant.kiev.zoral.com.ua> <20070712225324.F9515@besplex.bde.org> <20070712142127.GD2200@deviant.kiev.zoral.com.ua> <20070716195556.P12807@besplex.bde.org> <20070721063434.GI2200@deviant.kiev.zoral.com.ua> <20070721233613.Q3366@besplex.bde.org> <20070804075730.GP2738@deviant.kiev.zoral.com.ua> <20070808004001.O926@besplex.bde.org> <20070807170259.GJ2738@deviant.kiev.zoral.com.ua> <20070810133946.H769@besplex.bde.org> <20070810124153.GW2738@deviant.kiev.zoral.com.ua> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: bugs@FreeBSD.org, fs@FreeBSD.org Subject: Re: msdosfs not MPSAFE X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 14 Aug 2007 07:00:31 -0000 On Fri, 10 Aug 2007, Kostik Belousov wrote: > On Fri, Aug 10, 2007 at 01:54:48PM +1000, Bruce Evans wrote: >> I wrote yet another patch, with allocation on the stack so that no locking >> is required. This is simpler and doesn't require any new functions. >> Unfortunately, it is larger because it changes the interfaces for most >> functions. The interface changes are routine, so this is probably better. >> Note that 'struct dirent's are already allocated on the stack. This >> patch adds allocation of 'struct mbnambuf's which are slightly smaller >> (~256 bytes). I think this is just small enough for stack allocation. > > I agree that this is the best approach. The size of the on-stack > structure still make me worry, although ~270 bytes seems to be not too > large for 3-pages stack. Stack growth seems to be nowhere near a problem. With the extra ~270 bytes. ls -lR on i386 uses less than a 1-page stack (about 3.5K, including 0x270 bytes for the pcb). I think the maximum stack depth is attained when a debugger trap traps a fast interrupt interrupting a page fault in an i/o routine called from msdosfs_readdir(). Unlike in RELENG_4, nested interrupts can't occur, so a 1-page stack should be enough for everything on i386 (but of course isn't quite enough). I didn't try hard to attain the maximum depth, and just looked at where the stack got to after running a large ls -lR for a while. msdosfs_readdir() now allocates 0x2b0 bytes on the stack using "subl $0x2b0,%esp" and that is now the largest single allocation. This is without INVARIANTS etc. Bruce From owner-freebsd-fs@FreeBSD.ORG Tue Aug 14 09:38:35 2007 Return-Path: Delivered-To: fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id CADBC16A41A; Tue, 14 Aug 2007 09:38:35 +0000 (UTC) (envelope-from jroberson@chesapeake.net) Received: from webaccess-cl.virtdom.com (webaccess-cl.virtdom.com [216.240.101.25]) by mx1.freebsd.org (Postfix) with ESMTP id 7DAB013C461; Tue, 14 Aug 2007 09:38:35 +0000 (UTC) (envelope-from jroberson@chesapeake.net) Received: from [192.168.1.103] (c-71-231-138-78.hsd1.or.comcast.net [71.231.138.78]) (authenticated bits=0) by webaccess-cl.virtdom.com (8.13.6/8.13.6) with ESMTP id l7E9cVMs039213 (version=TLSv1/SSLv3 cipher=DHE-DSS-AES256-SHA bits=256 verify=NO); Tue, 14 Aug 2007 05:38:32 -0400 (EDT) (envelope-from jroberson@chesapeake.net) Date: Tue, 14 Aug 2007 02:41:31 -0700 (PDT) From: Jeff Roberson X-X-Sender: jroberson@10.0.0.1 To: fs@freebsd.org, net@freebsd.org Message-ID: <20070814024024.U568@10.0.0.1> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: Subject: netncp/netsmb users please test a patch. X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 14 Aug 2007 09:38:35 -0000 http://people.freebsd.org/~jeff/select.diff I have redone the select locking. This included changing some cruft in smb/ncp. I have tested smb myself, but would appreciate more feedback. I am not able to test ncp. Please let me know if this works for you. Thanks, Jeff From owner-freebsd-fs@FreeBSD.ORG Tue Aug 14 20:33:27 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6D3F716A418 for ; Tue, 14 Aug 2007 20:33:27 +0000 (UTC) (envelope-from gore_jarold@yahoo.com) Received: from web63008.mail.re1.yahoo.com (web63008.mail.re1.yahoo.com [69.147.96.219]) by mx1.freebsd.org (Postfix) with SMTP id 107CB13C45A for ; Tue, 14 Aug 2007 20:33:26 +0000 (UTC) (envelope-from gore_jarold@yahoo.com) Received: (qmail 85742 invoked by uid 60001); 14 Aug 2007 20:33:26 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=n01ToqrbyOhkD/CLucNP1hP+HkVX96d8j3ikyNOb+fpxuFEGtZkYlR+Kxz4xkbRP+9y8T33esBKbMyw9xP6oqqUBQzsoJdbzcRPzG4RnlKBR/cDyoJ+qxhgTAHe7cdkHV01WDTEBKGuGpt2IyfQ9pPHTlPmnxMKv+JbXTLYCp54=; X-YMail-OSG: xkpstQ0VM1mhBh0SFeWQrL9kX5n5C3kjUY7HI4ezGTIj4h_3lJ.4hUbHnSM0Ow3iemQFs.7tsfFU23x4ztuzf9Dx8XazxGq.yqYMx1X8dx4rD9b_QsmJRz10lPH6VI9A Received: from [71.63.232.32] by web63008.mail.re1.yahoo.com via HTTP; Tue, 14 Aug 2007 13:33:26 PDT Date: Tue, 14 Aug 2007 13:33:26 -0700 (PDT) From: Gore Jarold To: freebsd-fs@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <186173.85039.qm@web63008.mail.re1.yahoo.com> Cc: ivoras@fer.hr Subject: Re: help needed - tuning a filesystem for rm and cp ? (MORE) X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 14 Aug 2007 20:33:27 -0000 On Fri, 13 Jul 2007, Ivan Voras wrote: > Gore Jarold wrote: > > > vfs.ufs.dirhash_maxmem: 2097152 > > vfs.ufs.dirhash_mem: 2065716 > > > > > > Interesting at all ? > > Yes, you've used up all dirhash_mem. Since you have enough memory, try > increasing dirhash_maxmem by a factor of 4 or more and testing again. It > might help you with large directories (lots of files). Ok, you are correct - I am indeed maxing out vfs.ufs.dirhash_mem I have just upped it by 2x and will continue to monitor it. Here is a question for any and all out there reading ... what would you expect would happen to a system that was constantly maxing out this value, sometimes on a sustained basis, while the activity that caused it went on uninterrupted ? I am seeing the system halt ... is it reasonable to think that maxing that value out on a regular, sustained basis would cause a system to halt ? (6.2-release running on a 4 GB memory p4 xeon ... does nothing but fileserver duties) ____________________________________________________________________________________ Need a vacation? Get great deals to amazing places on Yahoo! Travel. http://travel.yahoo.com/ From owner-freebsd-fs@FreeBSD.ORG Tue Aug 14 20:44:57 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D9A4C16A420 for ; Tue, 14 Aug 2007 20:44:57 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.ltdomains.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id B039413C46A for ; Tue, 14 Aug 2007 20:44:57 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from proton.local (209-163-168-124.static.twtelecom.net [209.163.168.124]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l7EKirIr027006; Tue, 14 Aug 2007 15:44:54 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <46C21446.40004@freebsd.org> Date: Tue, 14 Aug 2007 15:44:54 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.6 (Macintosh/20070728) MIME-Version: 1.0 To: Gore Jarold References: <186173.85039.qm@web63008.mail.re1.yahoo.com> In-Reply-To: <186173.85039.qm@web63008.mail.re1.yahoo.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-1.5 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: freebsd-fs@freebsd.org, ivoras@fer.hr Subject: Re: help needed - tuning a filesystem for rm and cp ? (MORE) X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 14 Aug 2007 20:44:57 -0000 Gore Jarold wrote: > On Fri, 13 Jul 2007, Ivan Voras wrote: > >> Gore Jarold wrote: >> >>> vfs.ufs.dirhash_maxmem: 2097152 >>> vfs.ufs.dirhash_mem: 2065716 >>> >>> >>> Interesting at all ? >> Yes, you've used up all dirhash_mem. Since you have > enough memory, try >> increasing dirhash_maxmem by a factor of 4 or more > and testing again. It >> might help you with large directories (lots of > files). > > > Ok, you are correct - I am indeed maxing out > vfs.ufs.dirhash_mem > > I have just upped it by 2x and will continue to > monitor it. > > Here is a question for any and all out there reading > ... what would you expect would happen to a system > that was constantly maxing out this value, sometimes > on a sustained basis, while the activity that caused > it went on uninterrupted ? > > I am seeing the system halt ... is it reasonable to > think that maxing that value out on a regular, > sustained basis would cause a system to halt ? > > (6.2-release running on a 4 GB memory p4 xeon ... does > nothing but fileserver duties) If you have a lot of meta-data IO (which you seem to have), then it's possible that the system is incredibly busy doing disk accesses, and waiting on IO from storage. When you say 'halt' does that mean you can't log in to it, and eventually it comes back alive, or does that mean it is locked up in a way that never recovers? Eric From owner-freebsd-fs@FreeBSD.ORG Tue Aug 14 21:07:59 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 7F2B216A469 for ; Tue, 14 Aug 2007 21:07:59 +0000 (UTC) (envelope-from speedtoys.racing@gmail.com) Received: from py-out-1112.google.com (py-out-1112.google.com [64.233.166.176]) by mx1.freebsd.org (Postfix) with ESMTP id 3E5C513C4B3 for ; Tue, 14 Aug 2007 21:07:59 +0000 (UTC) (envelope-from speedtoys.racing@gmail.com) Received: by py-out-1112.google.com with SMTP id a73so3270524pye for ; Tue, 14 Aug 2007 14:07:58 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:references:message-id:from:to:in-reply-to:content-type:x-mailer:mime-version:subject:content-transfer-encoding:date:cc; b=L8LHuL2rl6jA+aYtZtOWYUpEb/Baw7NKGqhZc6CQFB4ZwmJJ6XN/7YKPNEnEWZMwX0pfgEXa+o5tGcMlJP2yPIpu3Rct/XMq+5Qw31j397yMGcmplM6sXWGh1O+S1n7y1CpDf/bhBokAljKNP951d/ogyWK2kCVEEZA+QYPEff4= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:references:message-id:from:to:in-reply-to:content-type:x-mailer:mime-version:subject:content-transfer-encoding:date:cc; b=GJDVrAX2Vj+Glcpk6S4dW7i3TEqjqCvHRxOc9oU2laQdbX5tuHXcagI44tXbgmTZiddW0fY9MCaPj1xdlz1M3KIFc4YN8CcczjPeT5uKCyQFQzqlxUKkIIbyLNDogszKHtkF4S0tRPIldsKsaFjvhR0igIbFsGSpkpABOd/PJ4o= Received: by 10.35.129.19 with SMTP id g19mr10762506pyn.1187124160613; Tue, 14 Aug 2007 13:42:40 -0700 (PDT) Received: from ?10.24.173.178? ( [32.168.149.26]) by mx.google.com with ESMTPS id v55sm8169796pyh.2007.08.14.13.42.36 (version=TLSv1/SSLv3 cipher=OTHER); Tue, 14 Aug 2007 13:42:38 -0700 (PDT) References: <186173.85039.qm@web63008.mail.re1.yahoo.com> Message-Id: <14087D0D-B8A7-44F9-9EA7-2DA6C0DEB733@gmail.com> From: Speedtoys To: Gore Jarold In-Reply-To: <186173.85039.qm@web63008.mail.re1.yahoo.com> Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes X-Mailer: iPhone Mail (1C25) Mime-Version: 1.0 (iPhone Mail 1C25) Content-Transfer-Encoding: 7bit Date: Tue, 14 Aug 2007 13:42:26 -0700 Cc: "freebsd-fs@freebsd.org" , "ivoras@fer.hr" Subject: Re: help needed - tuning a filesystem for rm and cp ? (MORE) X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 14 Aug 2007 21:07:59 -0000 Will upping these values make a difference on nfs volumes as well? Got brakes? ====== 25hrs or one season with one pad set is possible. Save money and pit time, compromise nothing. Ask how. TXT or Tone: 8414546712@txt.att.net http://www.speedtoys.com On Aug 14, 2007, at 1:33 PM, Gore Jarold wrote: > On Fri, 13 Jul 2007, Ivan Voras wrote: > >> Gore Jarold wrote: >> >>> vfs.ufs.dirhash_maxmem: 2097152 >>> vfs.ufs.dirhash_mem: 2065716 >>> >>> >>> Interesting at all ? >> >> Yes, you've used up all dirhash_mem. Since you have > enough memory, try >> increasing dirhash_maxmem by a factor of 4 or more > and testing again. It >> might help you with large directories (lots of > files). > > > Ok, you are correct - I am indeed maxing out > vfs.ufs.dirhash_mem > > I have just upped it by 2x and will continue to > monitor it. > > Here is a question for any and all out there reading > ... what would you expect would happen to a system > that was constantly maxing out this value, sometimes > on a sustained basis, while the activity that caused > it went on uninterrupted ? > > I am seeing the system halt ... is it reasonable to > think that maxing that value out on a regular, > sustained basis would cause a system to halt ? > > (6.2-release running on a 4 GB memory p4 xeon ... does > nothing but fileserver duties) > > > > ____________________________________________________________________________________ > Need a vacation? Get great deals > to amazing places on Yahoo! Travel. > http://travel.yahoo.com/ > _______________________________________________ > freebsd-fs@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-fs > To unsubscribe, send any mail to "freebsd-fs-unsubscribe@freebsd.org" From owner-freebsd-fs@FreeBSD.ORG Tue Aug 14 21:20:39 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E15E216A419 for ; Tue, 14 Aug 2007 21:20:39 +0000 (UTC) (envelope-from freebsd-fs@m.gmane.org) Received: from ciao.gmane.org (main.gmane.org [80.91.229.2]) by mx1.freebsd.org (Postfix) with ESMTP id 6D42213C4A3 for ; Tue, 14 Aug 2007 21:20:39 +0000 (UTC) (envelope-from freebsd-fs@m.gmane.org) Received: from list by ciao.gmane.org with local (Exim 4.43) id 1IL3oR-0006NZ-Eq for freebsd-fs@freebsd.org; Tue, 14 Aug 2007 23:20:27 +0200 Received: from 78-1-115-156.adsl.net.t-com.hr ([78.1.115.156]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 14 Aug 2007 23:20:27 +0200 Received: from ivoras by 78-1-115-156.adsl.net.t-com.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 14 Aug 2007 23:20:27 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-fs@freebsd.org From: Ivan Voras Date: Tue, 14 Aug 2007 23:20:07 +0200 Lines: 47 Message-ID: References: <186173.85039.qm@web63008.mail.re1.yahoo.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enig8C3866F7E859A88A638A6E66" X-Complaints-To: usenet@sea.gmane.org X-Gmane-NNTP-Posting-Host: 78-1-115-156.adsl.net.t-com.hr User-Agent: Thunderbird 1.5.0.12 (Windows/20070509) In-Reply-To: <186173.85039.qm@web63008.mail.re1.yahoo.com> X-Enigmail-Version: 0.94.3.0 Sender: news Subject: Re: help needed - tuning a filesystem for rm and cp ? (MORE) X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 14 Aug 2007 21:20:40 -0000 This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig8C3866F7E859A88A638A6E66 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Gore Jarold wrote: > Ok, you are correct - I am indeed maxing out > vfs.ufs.dirhash_mem >=20 > I have just upped it by 2x and will continue to > monitor it. >=20 > Here is a question for any and all out there reading > ... what would you expect would happen to a system > that was constantly maxing out this value, sometimes > on a sustained basis, while the activity that caused > it went on uninterrupted ? A slowdown. Dirhash speeds up file access when the number of files is large (i.e. large directories). It's not required for proper function. > I am seeing the system halt ... is it reasonable to > think that maxing that value out on a regular, > sustained basis would cause a system to halt ? No, system halting or crashing points to a bug. If it's really halted ("freezed", no network or IO), someone should be able to lead you through a debugging session. --------------enig8C3866F7E859A88A638A6E66 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.5 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGwhyNldnAQVacBcgRAiC+AKCBF1bBIOTOpA/EinawfHHSSK7CrwCdHQaM X4HkGVnpq0y7d9ptmN9yWWM= =O8wr -----END PGP SIGNATURE----- --------------enig8C3866F7E859A88A638A6E66-- From owner-freebsd-fs@FreeBSD.ORG Tue Aug 14 21:47:18 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 4339C16A421 for ; Tue, 14 Aug 2007 21:47:18 +0000 (UTC) (envelope-from gore_jarold@yahoo.com) Received: from web63006.mail.re1.yahoo.com (web63006.mail.re1.yahoo.com [69.147.96.217]) by mx1.freebsd.org (Postfix) with SMTP id F2FB313C45D for ; Tue, 14 Aug 2007 21:47:17 +0000 (UTC) (envelope-from gore_jarold@yahoo.com) Received: (qmail 71886 invoked by uid 60001); 14 Aug 2007 21:47:16 -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=js9XTknRCcgNXt3uYjiutkWXTidJNr1scy5lbsvPg6zQjyDPC8lyzTtFwweFAH6SAWf52Wfl+5BfXKo2+JXj5q27nIdzeHPSpbaCc/e7NsPOK0fUw8jpxVSVZjVzwe4jNPxMFALolyF11p8W8uWl2dPyzet3DDnqXbIityEywT0=; X-YMail-OSG: x3V68gwVM1lOmzVw58yuDpqtFtI4aLAUthwJuhSuVzfPoSxhzVcr5Y_HTcgr4R5y3KAZ8rPTEnsg8gxAp_PP0VKeEAGsXv1AK_j.LQf5uF49.Twr.5235I8loQGENPG. Received: from [71.63.232.32] by web63006.mail.re1.yahoo.com via HTTP; Tue, 14 Aug 2007 14:47:16 PDT Date: Tue, 14 Aug 2007 14:47:16 -0700 (PDT) From: Gore Jarold To: freebsd-fs@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <924334.71021.qm@web63006.mail.re1.yahoo.com> Subject: Re: help needed - tuning a filesystem for rm and cp ? (more details) X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 14 Aug 2007 21:47:18 -0000 > > Here is a question for any and all out there reading > > ... what would you expect would happen to a system > > that was constantly maxing out this value, sometimes > > on a sustained basis, while the activity that caused > > it went on uninterrupted ? > > > > I am seeing the system halt ... is it reasonable to > > think that maxing that value out on a regular, > > sustained basis would cause a system to halt ? > > > > (6.2-release running on a 4 GB memory p4 xeon ... does > > nothing but fileserver duties) > > > If you have a lot of meta-data IO (which you seem to have), then it's > possible that the system is incredibly busy doing disk accesses, and > waiting on IO from storage. When you say 'halt' does that mean you > can't log in to it, and eventually it comes back alive, or does that > mean it is locked up in a way that never recovers? I never wait around to find out it it comes back ... It no longer responds to pings, so I assume it is actually crashed. Also, it should be noted that this system (and its large, >4TB filesystem) has quotas enabled. So perhaps a better question would be: Any comments on frequent, sustained maxing out of dirhash on a quota'd filesystem ? ____________________________________________________________________________________Ready for the edge of your seat? Check out tonight's top picks on Yahoo! TV. http://tv.yahoo.com/ From owner-freebsd-fs@FreeBSD.ORG Wed Aug 15 03:24:03 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 50B1516A420 for ; Wed, 15 Aug 2007 03:24:03 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.ltdomains.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id 26BFD13C480 for ; Wed, 15 Aug 2007 03:24:03 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from neutrino.vnode.org (r74-193-81-203.pfvlcmta01.grtntx.tl.dh.suddenlink.net [74.193.81.203]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l7F3O13C043380 (version=TLSv1/SSLv3 cipher=DHE-DSS-AES256-SHA bits=256 verify=NO); Tue, 14 Aug 2007 22:24:02 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <46C271CC.8080008@freebsd.org> Date: Tue, 14 Aug 2007 22:23:56 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.6 (X11/20070812) MIME-Version: 1.0 To: Gore Jarold References: <924334.71021.qm@web63006.mail.re1.yahoo.com> In-Reply-To: <924334.71021.qm@web63006.mail.re1.yahoo.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-1.4 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: freebsd-fs@freebsd.org Subject: Re: help needed - tuning a filesystem for rm and cp ? (more details) X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 15 Aug 2007 03:24:03 -0000 On 08/14/07 16:47, Gore Jarold wrote: >>> Here is a question for any and all out there > reading >>> ... what would you expect would happen to a system >>> that was constantly maxing out this value, > sometimes >>> on a sustained basis, while the activity that > caused >>> it went on uninterrupted ? >>> >>> I am seeing the system halt ... is it reasonable > to >>> think that maxing that value out on a regular, >>> sustained basis would cause a system to halt ? >>> >>> (6.2-release running on a 4 GB memory p4 xeon ... > does >>> nothing but fileserver duties) >> >> If you have a lot of meta-data IO (which you seem to > have), then it's >> possible that the system is incredibly busy doing > disk accesses, and >> waiting on IO from storage. When you say 'halt' > does that mean you >> can't log in to it, and eventually it comes back > alive, or does that >> mean it is locked up in a way that never recovers? > > > I never wait around to find out it it comes back ... > > It no longer responds to pings, so I assume it is > actually crashed. > > Also, it should be noted that this system (and its > large, >4TB filesystem) has quotas enabled. > > So perhaps a better question would be: > > Any comments on frequent, sustained maxing out of > dirhash on a quota'd filesystem ? If you can, it would be best if you can break into the debugger and get a core. I can't think of how dirhash would effect quotas really, but I suppose it might be possible. You might try bumping the dirhash setting way up, maybe to 20MB or so. I think it might be worthwhile also to run a script (cron maybe) every minute that logs the current use of dirhash, and maybe a ps -auxwl, with a date stamp so you can when your peaks occur, what was running, etc, just before a crash. Might also look into bsdsar - it might help with some of this too. Eric From owner-freebsd-fs@FreeBSD.ORG Wed Aug 15 12:21:50 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 01EBF16A417 for ; Wed, 15 Aug 2007 12:21:50 +0000 (UTC) (envelope-from scsvc1@scsupport.net) Received: from camel.he.net (camel.he.net [216.218.242.2]) by mx1.freebsd.org (Postfix) with SMTP id E2B6A13C467 for ; Wed, 15 Aug 2007 12:21:49 +0000 (UTC) (envelope-from scsvc1@scsupport.net) Message-Id: <1187170186.18612@camel.he.net> Date: Wed, 15 Aug 2007 02:29:46 -0700 To: freebsd-fs@freebsd.org From: SwissCash Support Content-Transfer-Encoding: 8bit MIME-Version: 1.0 Content-Type: text/plain X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Subject: [SwissCash Warning] Multiple password failure X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: scsvc1@scsupport.net List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 15 Aug 2007 12:21:50 -0000 Dear Investors, We recently have determined that different computers have logged into your SwissCash Account, and multiple password failure were present before the log on. We need you to re-confirm your account information with us now. If this is not completed we will be forced to suspend your account indefinitely. We thank you for your cooperation in this manner. To confirm your account records click here : [1]https://secured.sip25.com/web/login.aspx [2]https://secured.swisscash.net/web/login.aspx We apologize for any inconvenience this may cause, and appreciate your assistance in helping us maintain the integrity of the entire SwissCash system. We thank you for your prompt attention to this matter. Please understand that this is a security measure intended to help protect us and your account. Thank you, We are here always to serve you better. Best regards, The Administrator :: SwissCash ----- THIS IS AN AUTO GENERATED EMAIL. PLEASE DO NOT REPLY ----- References 1. http://secured.sip25.bz/web/login.aspx/ 2. http://secured.swisscash.net.in/web/login.aspx/ From owner-freebsd-fs@FreeBSD.ORG Wed Aug 15 16:27:12 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E86AA16A418 for ; Wed, 15 Aug 2007 16:27:12 +0000 (UTC) (envelope-from gore_jarold@yahoo.com) Received: from web63012.mail.re1.yahoo.com (web63012.mail.re1.yahoo.com [69.147.96.223]) by mx1.freebsd.org (Postfix) with SMTP id 837DD13C481 for ; Wed, 15 Aug 2007 16:27:12 +0000 (UTC) (envelope-from gore_jarold@yahoo.com) Received: (qmail 54065 invoked by uid 60001); 15 Aug 2007 16:27:11 -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=uJ6iwKhz1D7PAFN2sVuPT80nGoYJwHvXNCa3jFuf4ahfutQmLtbrvnHWJUUPqpty/BligqypLSIIq1ffd8CVEHqb2KDYxsuJ+18VNmtAXgDlAOlIwXDJH98CPbEFY57qBe6Iib9LnIxx7nrdv45Ttyc9TA8iR96DSES08zHBTtU=; X-YMail-OSG: FZQZWusVM1lsRtkNuDcjN6_uiD7vsPY4zeFM6W3ulm6O9rp5N0Fb.7JxtA9l50J8xw-- Received: from [71.63.232.32] by web63012.mail.re1.yahoo.com via HTTP; Wed, 15 Aug 2007 09:27:11 PDT Date: Wed, 15 Aug 2007 09:27:11 -0700 (PDT) From: Gore Jarold To: freebsd-fs@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <723681.52797.qm@web63012.mail.re1.yahoo.com> Subject: vfs.ufs.dirhash_maxmem ... how high can I go ? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 15 Aug 2007 16:27:13 -0000 Based on the previous discussion, I have decided to up my dirhash_maxmem and also to start tracking its usage over time ... I doubled it, then doubled again, then again ... and the system even hit the 16 MB ceiling. So now I have set it to 32 MB and we will see if the system can max that out. In the meantime, are there any unexpected consequences of upping this value to 32 MB ? I have 4 GB of physical and 4 GB of swap, running on normal 32-bit x86, and I have this set as well: kern.maxdsiz="2572000000" any comments ? Also, now that we see that my dirhash usage is _very high_, any revised comments on the ability of these workloads to crash a machine that was at the 2 MB default ? ____________________________________________________________________________________ Got a little couch potato? Check out fun summer activities for kids. http://search.yahoo.com/search?fr=oni_on_mail&p=summer+activities+for+kids&cs=bz From owner-freebsd-fs@FreeBSD.ORG Wed Aug 15 19:27:20 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 228CB16A419 for ; Wed, 15 Aug 2007 19:27:20 +0000 (UTC) (envelope-from wilson.david.l@gmail.com) Received: from wx-out-0506.google.com (wx-out-0506.google.com [66.249.82.236]) by mx1.freebsd.org (Postfix) with ESMTP id D428F13C457 for ; Wed, 15 Aug 2007 19:27:19 +0000 (UTC) (envelope-from wilson.david.l@gmail.com) Received: by wx-out-0506.google.com with SMTP id i29so16424wxd for ; Wed, 15 Aug 2007 12:27:19 -0700 (PDT) DKIM-Signature: a=rsa-sha1; c=relaxed/relaxed; d=gmail.com; s=beta; h=domainkey-signature:received:received:message-id:date:from:to:subject:mime-version:content-type; b=MAoEDBcl0UhcyrWdAkqnJ8MBtaX3WRePujKas2EQDH7BwPsByVEjcJEsiCbgZqN0RiyibGN/y1g9pY3ZL6Tm2it07C4jSJam1HXF64LaRP5UEgTu7LyI+EW9rTrqqW1REMKkSjbZQF05eduy2iytVjKTwpQ7NPs3IhHgzXTpvtk= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=beta; h=received:message-id:date:from:to:subject:mime-version:content-type; b=XyXBKoiS01SQnBLnzUpRXlZYmKTcKazs37cHB/C2ywNjSa81NGacfzH0FPR2UHps/JaPaHL+WQcrLs8htwcyx34OLoW+7ucqhboCMXP6a8aktSkHTsuyJD86Q0dbAue3UR2JScbWdYLUNWV/Tc80x8jPB3JW6xqmotEHFWEW7i8= Received: by 10.70.98.17 with SMTP id v17mr1414953wxb.1187204518463; Wed, 15 Aug 2007 12:01:58 -0700 (PDT) Received: by 10.70.94.17 with HTTP; Wed, 15 Aug 2007 12:01:58 -0700 (PDT) Message-ID: <3d51aded0708151201o72d6bf3mf1029545983180f6@mail.gmail.com> Date: Wed, 15 Aug 2007 15:01:58 -0400 From: "David Wilson" To: freebsd-fs@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Subject: mkuzip Utility X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 15 Aug 2007 19:27:20 -0000 Hello All, I have scoured the documents and searched the web for hours but cannot find an explanation of the steps needed to build a .uzip file using the mkuzip utility. Can anyone please provide me with instructions on using this tool? Thanks. From owner-freebsd-fs@FreeBSD.ORG Wed Aug 15 22:05:16 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 40FA016A418 for ; Wed, 15 Aug 2007 22:05:16 +0000 (UTC) (envelope-from des@des.no) Received: from tim.des.no (tim.des.no [194.63.250.121]) by mx1.freebsd.org (Postfix) with ESMTP id F1F7613C461 for ; Wed, 15 Aug 2007 22:05:15 +0000 (UTC) (envelope-from des@des.no) Received: from tim.des.no (localhost [127.0.0.1]) by spam.des.no (Postfix) with ESMTP id 606B320B0; Wed, 15 Aug 2007 23:48:15 +0200 (CEST) X-Spam-Tests: AWL X-Spam-Learn: disabled X-Spam-Score: 0.0/3.0 X-Spam-Checker-Version: SpamAssassin 3.2.1 (2007-05-02) on tim.des.no Received: from ds4.des.no (des.no [80.203.243.180]) by smtp.des.no (Postfix) with ESMTP id CE37B208A; Wed, 15 Aug 2007 23:48:14 +0200 (CEST) Received: by ds4.des.no (Postfix, from userid 1001) id AC9558444F; Wed, 15 Aug 2007 23:48:14 +0200 (CEST) From: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= To: "David Wilson" References: <3d51aded0708151201o72d6bf3mf1029545983180f6@mail.gmail.com> Date: Wed, 15 Aug 2007 23:48:14 +0200 In-Reply-To: <3d51aded0708151201o72d6bf3mf1029545983180f6@mail.gmail.com> (David Wilson's message of "Wed\, 15 Aug 2007 15\:01\:58 -0400") Message-ID: <86zm0spizl.fsf@ds4.des.no> User-Agent: Gnus/5.110006 (No Gnus v0.6) Emacs/22.1 (berkeley-unix) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Cc: freebsd-fs@freebsd.org Subject: Re: mkuzip Utility X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 15 Aug 2007 22:05:16 -0000 "David Wilson" writes: > I have scoured the documents and searched the web for hours but cannot fi= nd > an explanation of the steps needed to build a .uzip file using the mkuzip > utility. Can anyone please provide me with instructions on using this to= ol? The man page pretty much says it all. The input is a disk image, either copied off a physical disk or created from scratch. DES --=20 Dag-Erling Sm=C3=B8rgrav - des@des.no From owner-freebsd-fs@FreeBSD.ORG Thu Aug 16 10:37:20 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 0488316A46C for ; Thu, 16 Aug 2007 10:37:20 +0000 (UTC) (envelope-from lulf@stud.ntnu.no) Received: from fri.itea.ntnu.no (fri.itea.ntnu.no [129.241.7.60]) by mx1.freebsd.org (Postfix) with ESMTP id 6BC6513C465 for ; Thu, 16 Aug 2007 10:37:18 +0000 (UTC) (envelope-from lulf@stud.ntnu.no) Received: from localhost (localhost [127.0.0.1]) by fri.itea.ntnu.no (Postfix) with ESMTP id A1E3199D7 for ; Thu, 16 Aug 2007 12:06:08 +0200 (CEST) Received: from caracal.stud.ntnu.no (caracal.stud.ntnu.no [129.241.56.185]) by fri.itea.ntnu.no (Postfix) with ESMTP for ; Thu, 16 Aug 2007 12:05:26 +0200 (CEST) Received: by caracal.stud.ntnu.no (Postfix, from userid 2312) id D71D96240E2; Thu, 16 Aug 2007 12:05:26 +0200 (CEST) Date: Thu, 16 Aug 2007 12:05:26 +0200 From: Ulf Lilleengen To: freebsd-fs@freebsd.org Message-ID: <20070816100526.GA31897@stud.ntnu.no> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="7AUc2qLy4jB3hD7Z" Content-Disposition: inline User-Agent: Mutt/1.5.9i X-Content-Scanned: with sophos and spamassassin at mailgw.ntnu.no. Subject: [PATCH] Make fdescfs MPSAFE X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 16 Aug 2007 10:37:20 -0000 --7AUc2qLy4jB3hD7Z Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Hi, To be able to better understand VFS and locking in general, I started making fdescfs MPSAFE. I'm not experienced with any of these things, so there might be some errors, although I've looked through much VFS code and code for other FS like nullfs. I've tested it by running two pthreads on the same fd, and that seamt to work, but there might be other cases where it will fail. Patch is attached. -- Ulf Lilleengen --7AUc2qLy4jB3hD7Z Content-Type: text/plain; charset=us-ascii Content-Disposition: attachment; filename="fdescfs_locking.diff" Index: sys/fs/fdescfs/fdesc.h =================================================================== RCS file: /home/ncvs/src/sys/fs/fdescfs/fdesc.h,v retrieving revision 1.20 diff -u -r1.20 fdesc.h --- sys/fs/fdescfs/fdesc.h 10 Feb 2005 12:09:15 -0000 1.20 +++ sys/fs/fdescfs/fdesc.h 13 Aug 2007 23:29:21 -0000 @@ -31,7 +31,7 @@ * * @(#)fdesc.h 8.5 (Berkeley) 1/21/94 * - * $FreeBSD: src/sys/fs/fdescfs/fdesc.h,v 1.20 2005/02/10 12:09:15 phk Exp $ + * $FreeBSD$ */ #ifdef _KERNEL @@ -59,6 +59,7 @@ #define VTOFDESC(vp) ((struct fdescnode *)(vp)->v_data) extern vfs_init_t fdesc_init; +extern vfs_uninit_t fdesc_uninit; extern int fdesc_allocvp(fdntype, int, struct mount *, struct vnode **, struct thread *); #endif /* _KERNEL */ Index: sys/fs/fdescfs/fdesc_vfsops.c =================================================================== RCS file: /home/ncvs/src/sys/fs/fdescfs/fdesc_vfsops.c,v retrieving revision 1.56 diff -u -r1.56 fdesc_vfsops.c --- sys/fs/fdescfs/fdesc_vfsops.c 4 Apr 2007 09:11:32 -0000 1.56 +++ sys/fs/fdescfs/fdesc_vfsops.c 13 Aug 2007 23:29:21 -0000 @@ -31,7 +31,7 @@ * * @(#)fdesc_vfsops.c 8.4 (Berkeley) 1/21/94 * - * $FreeBSD: src/sys/fs/fdescfs/fdesc_vfsops.c,v 1.56 2007/04/04 09:11:32 rwatson Exp $ + * $FreeBSD$ */ /* @@ -85,17 +85,22 @@ if (mp->mnt_flag & (MNT_UPDATE | MNT_ROOTFS)) return (EOPNOTSUPP); - error = fdesc_allocvp(Froot, FD_ROOT, mp, &rvp, td); - if (error) - return (error); - MALLOC(fmp, struct fdescmount *, sizeof(struct fdescmount), M_FDESCMNT, M_WAITOK); /* XXX */ + error = fdesc_allocvp(Froot, FD_ROOT, mp, &rvp, td); + if (error) { + free(fmp, M_FDESCMNT); + return (error); + } rvp->v_type = VDIR; rvp->v_vflag |= VV_ROOT; fmp->f_root = rvp; + VOP_UNLOCK(rvp, 0, td); /* XXX -- don't mark as local to work around fts() problems */ /*mp->mnt_flag |= MNT_LOCAL;*/ + MNT_ILOCK(mp); + mp->mnt_kern_flag |= MNTK_MPSAFE; + MNT_IUNLOCK(mp); mp->mnt_data = (qaddr_t) fmp; vfs_getnewfsid(mp); @@ -148,8 +153,8 @@ * Return locked reference to root. */ vp = VFSTOFDESC(mp)->f_root; - VREF(vp); vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td); + VREF(vp); *vpp = vp; return (0); } @@ -209,6 +214,7 @@ .vfs_root = fdesc_root, .vfs_statfs = fdesc_statfs, .vfs_unmount = fdesc_unmount, + .vfs_uninit = fdesc_uninit, }; VFS_SET(fdesc_vfsops, fdescfs, VFCF_SYNTHETIC); Index: sys/fs/fdescfs/fdesc_vnops.c =================================================================== RCS file: /home/ncvs/src/sys/fs/fdescfs/fdesc_vnops.c,v retrieving revision 1.104 diff -u -r1.104 fdesc_vnops.c --- sys/fs/fdescfs/fdesc_vnops.c 4 Apr 2007 09:11:32 -0000 1.104 +++ sys/fs/fdescfs/fdesc_vnops.c 13 Aug 2007 23:29:21 -0000 @@ -31,7 +31,7 @@ * * @(#)fdesc_vnops.c 8.9 (Berkeley) 1/21/94 * - * $FreeBSD: src/sys/fs/fdescfs/fdesc_vnops.c,v 1.104 2007/04/04 09:11:32 rwatson Exp $ + * $FreeBSD$ */ /* @@ -58,7 +58,7 @@ #define FDL_WANT 0x01 #define FDL_LOCKED 0x02 -static int fdcache_lock; +static struct mtx fdesc_hashmtx; #define NFDCACHE 4 #define FD_NHASH(ix) \ @@ -97,9 +97,32 @@ { fdhashtbl = hashinit(NFDCACHE, M_CACHE, &fdhash); + mtx_init(&fdesc_hashmtx, "fdeschs", NULL, MTX_DEF); return (0); } +/* Cleanup. */ +int +fdesc_uninit(vfsp) + struct vfsconf *vfsp; +{ + + mtx_destroy(&fdesc_hashmtx); + return (0); +} + +static void +fdesc_insmntque_dtr(struct vnode *vp, void *arg) +{ + struct fdescnode *fd; + + fd = (struct fdescnode *)arg; + free(fd, M_TEMP); + vp->v_data = NULL; + vgone(vp); + vput(vp); +} + int fdesc_allocvp(ftype, ix, mp, vpp, td) fdntype ftype; @@ -114,25 +137,20 @@ fc = FD_NHASH(ix); loop: + mtx_lock(&fdesc_hashmtx); LIST_FOREACH(fd, fc, fd_hash) { if (fd->fd_ix == ix && fd->fd_vnode->v_mount == mp) { - if (vget(fd->fd_vnode, 0, td)) + VI_LOCK(fd->fd_vnode); + mtx_unlock(&fdesc_hashmtx); + if (vget(fd->fd_vnode, LK_EXCLUSIVE | LK_INTERLOCK, td)) { + VI_UNLOCK(fd->fd_vnode); goto loop; + } *vpp = fd->fd_vnode; return (error); } } - - /* - * otherwise lock the array while we call getnewvnode - * since that can block. - */ - if (fdcache_lock & FDL_LOCKED) { - fdcache_lock |= FDL_WANT; - (void) tsleep( &fdcache_lock, PINOD, "fdalvp", 0); - goto loop; - } - fdcache_lock |= FDL_LOCKED; + mtx_unlock(&fdesc_hashmtx); /* * Do the MALLOC before the getnewvnode since doing so afterward @@ -151,23 +169,14 @@ fd->fd_type = ftype; fd->fd_fd = -1; fd->fd_ix = ix; - /* XXX: vnode should be locked here */ - error = insmntque(*vpp, mp); /* XXX: Too early for mpsafe fs */ - if (error != 0) { - free(fd, M_TEMP); - *vpp = NULLVP; + vn_lock(*vpp, LK_EXCLUSIVE | LK_RETRY, td); + error = insmntque1(*vpp, mp, fdesc_insmntque_dtr, fd); + if (error != 0) goto out; - } + mtx_lock(&fdesc_hashmtx); LIST_INSERT_HEAD(fc, fd, fd_hash); - + mtx_unlock(&fdesc_hashmtx); out: - fdcache_lock &= ~FDL_LOCKED; - - if (fdcache_lock & FDL_WANT) { - fdcache_lock &= ~FDL_WANT; - wakeup( &fdcache_lock); - } - return (error); } @@ -233,8 +242,7 @@ if (error) goto bad; VTOFDESC(fvp)->fd_fd = fd; - if (fvp != dvp) - vn_lock(fvp, LK_EXCLUSIVE | LK_RETRY, td); + *vpp = fvp; return (0); @@ -528,7 +536,10 @@ struct vnode *vp = ap->a_vp; struct fdescnode *fd = VTOFDESC(vp); + mtx_lock(&fdesc_hashmtx); LIST_REMOVE(fd, fd_hash); + mtx_unlock(&fdesc_hashmtx); + FREE(vp->v_data, M_TEMP); vp->v_data = 0; --7AUc2qLy4jB3hD7Z-- From owner-freebsd-fs@FreeBSD.ORG Thu Aug 16 16:17:56 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8C1B616A417 for ; Thu, 16 Aug 2007 16:17:56 +0000 (UTC) (envelope-from gore_jarold@yahoo.com) Received: from web63005.mail.re1.yahoo.com (web63005.mail.re1.yahoo.com [69.147.96.216]) by mx1.freebsd.org (Postfix) with SMTP id 314BA13C442 for ; Thu, 16 Aug 2007 16:17:56 +0000 (UTC) (envelope-from gore_jarold@yahoo.com) Received: (qmail 63181 invoked by uid 60001); 16 Aug 2007 16:17:55 -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=g3/iY/8/YD1At1fSoY8LrUXSXJjSUKpfWkOZq6DvKrOAKKaqxG9kavFRwiff/wtFMiS5Xqwy0+11u791JHIcg7AWS+zv1wiJeQlN+P09U5tckbU+p3pq4NvbUk5sDH5dpmmgMvnvfCZXSAKtIc66bVscx37ZiEc1dU4LiTqMgwU=; X-YMail-OSG: VYQAPl0VM1lC27nG1I6Oml7cPHm23rz6SDF2S0AjIk4HaE6nmYYjmJLXYs4tsVxrSq0EBcqjwnRaWPErYgTetOr5hyM5Qk12IME- Received: from [71.63.232.32] by web63005.mail.re1.yahoo.com via HTTP; Thu, 16 Aug 2007 09:17:55 PDT Date: Thu, 16 Aug 2007 09:17:55 -0700 (PDT) From: Gore Jarold To: freebsd-fs@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <440077.63078.qm@web63005.mail.re1.yahoo.com> Subject: Re: vfs.ufs.dirhash_maxmem ... how high can I go ? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 16 Aug 2007 16:17:56 -0000 On Wed, 15 Aug 2007, Gore Jarold wrote: > Based on the previous discussion, I have decided to up > my dirhash_maxmem and also to start tracking its usage > over time ... > > I doubled it, then doubled again, then again ... and > the system even hit the 16 MB ceiling. > > So now I have set it to 32 MB and we will see if the > system can max that out. > > In the meantime, are there any unexpected consequences > of upping this value to 32 MB ? > > I have 4 GB of physical and 4 GB of swap, running on > normal 32-bit x86, and I have this set as well: > > kern.maxdsiz="2572000000" > > any comments ? > > Also, now that we see that my dirhash usage is _very > high_, any revised comments on the ability of these > workloads to crash a machine that was at the 2 MB > default ? Ok, it just ht the 32 MB limit. Bad things on the horizon if I up it further ? Or does my 4+4 GB memory and other details above make it benign ? ____________________________________________________________________________________ Take the Internet to Go: Yahoo!Go puts the Internet in your pocket: mail, news, photos & more. http://mobile.yahoo.com/go?refer=1GNXIC From owner-freebsd-fs@FreeBSD.ORG Thu Aug 16 17:13:48 2007 Return-Path: Delivered-To: freebsd-fs@FreeBSD.ORG Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8E77116A41B for ; Thu, 16 Aug 2007 17:13:48 +0000 (UTC) (envelope-from olli@lurza.secnetix.de) Received: from lurza.secnetix.de (lurza.secnetix.de [IPv6:2001:1b20:1:3::1]) by mx1.freebsd.org (Postfix) with ESMTP id E590413C45E for ; Thu, 16 Aug 2007 17:13:47 +0000 (UTC) (envelope-from olli@lurza.secnetix.de) Received: from lurza.secnetix.de (cdsdyb@localhost [127.0.0.1]) by lurza.secnetix.de (8.13.4/8.13.4) with ESMTP id l7GHDerg014550; Thu, 16 Aug 2007 19:13:46 +0200 (CEST) (envelope-from oliver.fromme@secnetix.de) Received: (from olli@localhost) by lurza.secnetix.de (8.13.4/8.13.1/Submit) id l7GHDexf014549; Thu, 16 Aug 2007 19:13:40 +0200 (CEST) (envelope-from olli) Date: Thu, 16 Aug 2007 19:13:40 +0200 (CEST) Message-Id: <200708161713.l7GHDexf014549@lurza.secnetix.de> From: Oliver Fromme To: freebsd-fs@FreeBSD.ORG, gore_jarold@yahoo.com In-Reply-To: <440077.63078.qm@web63005.mail.re1.yahoo.com> X-Newsgroups: list.freebsd-fs User-Agent: tin/1.8.2-20060425 ("Shillay") (UNIX) (FreeBSD/4.11-STABLE (i386)) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-2.1.2 (lurza.secnetix.de [127.0.0.1]); Thu, 16 Aug 2007 19:13:46 +0200 (CEST) Cc: Subject: Re: vfs.ufs.dirhash_maxmem ... how high can I go ? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: freebsd-fs@FreeBSD.ORG, gore_jarold@yahoo.com List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 16 Aug 2007 17:13:48 -0000 Gore Jarold wrote: > Gore Jarold wrote: > > Based on the previous discussion, I have decided > > to up my dirhash_maxmem and also to start tracking > > its usage over time ... > > > > I doubled it, then doubled again, then again ... > > and the system even hit the 16 MB ceiling. > > > > So now I have set it to 32 MB and we will see if > > the system can max that out. > > [...] > > Ok, it just ht the 32 MB limit. Well, you said you have > 20 million inodes. If all of them are accessed, I'm not surprised that your dirhash memory is maxed out. In that case FreeBSD will use up whatever amount you throw at it. Note that maxing it out is not a problem, it will just expire and overwrite old entries (it's only a kind of cache after all). You can even disable dirhash completely in your kernel, and the system should still run fine (albeit somewhat slower). > Bad things on the horizon if I up it further ? Or > does my 4+4 GB memory and other details above make > it benign ? Note that the dirhash memory is allocated from kernel memory, which is limited. See sysctl vm.kmem_size. And dirhash is not the only facility in the need of kernel memory, so you should be careful upping it any further. When the system runs out of kernel memory, it will panic. (However, you can increase the amount of kernel memory if necessary.) In any case, you should perform benchmarks with your typical application load, in order to find out if increasing dirhash_maxmem has any positive effect on it. If there's no clear performance improvement, it's not worth wasting kernel memory. Best regards Oliver -- Oliver Fromme, secnetix GmbH & Co. KG, Marktplatz 29, 85567 Grafing b. M. Handelsregister: Registergericht Muenchen, HRA 74606, Geschäftsfuehrung: secnetix Verwaltungsgesellsch. mbH, Handelsregister: Registergericht Mün- chen, HRB 125758, Geschäftsführer: Maik Bachmann, Olaf Erb, Ralf Gebhart FreeBSD-Dienstleistungen, -Produkte und mehr: http://www.secnetix.de/bsd "If you think C++ is not overly complicated, just what is a protected abstract virtual base pure virtual private destructor, and when was the last time you needed one?" -- Tom Cargil, C++ Journal From owner-freebsd-fs@FreeBSD.ORG Thu Aug 16 17:29:46 2007 Return-Path: Delivered-To: freebsd-fs@FreeBSD.ORG Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E75C316A469 for ; Thu, 16 Aug 2007 17:29:46 +0000 (UTC) (envelope-from olli@lurza.secnetix.de) Received: from lurza.secnetix.de (lurza.secnetix.de [IPv6:2001:1b20:1:3::1]) by mx1.freebsd.org (Postfix) with ESMTP id 4C05D13C474 for ; Thu, 16 Aug 2007 17:29:46 +0000 (UTC) (envelope-from olli@lurza.secnetix.de) Received: from lurza.secnetix.de (hmdazw@localhost [127.0.0.1]) by lurza.secnetix.de (8.13.4/8.13.4) with ESMTP id l7GHTdcS015169 for ; Thu, 16 Aug 2007 19:29:45 +0200 (CEST) (envelope-from oliver.fromme@secnetix.de) Received: (from olli@localhost) by lurza.secnetix.de (8.13.4/8.13.1/Submit) id l7GHTdb3015168; Thu, 16 Aug 2007 19:29:39 +0200 (CEST) (envelope-from olli) Date: Thu, 16 Aug 2007 19:29:39 +0200 (CEST) Message-Id: <200708161729.l7GHTdb3015168@lurza.secnetix.de> From: Oliver Fromme To: freebsd-fs@FreeBSD.ORG In-Reply-To: <14087D0D-B8A7-44F9-9EA7-2DA6C0DEB733@gmail.com> X-Newsgroups: list.freebsd-fs User-Agent: tin/1.8.2-20060425 ("Shillay") (UNIX) (FreeBSD/4.11-STABLE (i386)) MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-2.1.2 (lurza.secnetix.de [127.0.0.1]); Thu, 16 Aug 2007 19:29:45 +0200 (CEST) Cc: Subject: Re: help needed - tuning a filesystem for rm and cp ? (MORE) X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: freebsd-fs@FreeBSD.ORG List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 16 Aug 2007 17:29:47 -0000 Speedtoys wrote: > Will upping these values make a difference on nfs volumes as well? The dirhash optimization is for UFS only. It does not affect NFS clients. Of course, if your NFS _server_ exports any UFS file systems, it benefits from dirhash, too. Whether increasing the dirhash_max value will make a difference depends on two things. First check whether you're actually hitting the current maximum. If you don't, increasing it would be pointless. Second, if you did hit the maximum, try to increase it and perform some benchmarks to find out if the situation improves. If you can't measure a difference, I recommend to go back to the default value. Best regards Oliver PS: Please don't top-post. -- Oliver Fromme, secnetix GmbH & Co. KG, Marktplatz 29, 85567 Grafing b. M. Handelsregister: Registergericht Muenchen, HRA 74606, Geschäftsfuehrung: secnetix Verwaltungsgesellsch. mbH, Handelsregister: Registergericht Mün- chen, HRB 125758, Geschäftsführer: Maik Bachmann, Olaf Erb, Ralf Gebhart FreeBSD-Dienstleistungen, -Produkte und mehr: http://www.secnetix.de/bsd "One of the main causes of the fall of the Roman Empire was that, lacking zero, they had no way to indicate successful termination of their C programs." -- Robert Firth From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 01:28:52 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id A14D816A41A for ; Fri, 17 Aug 2007 01:28:52 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: from web34511.mail.mud.yahoo.com (web34511.mail.mud.yahoo.com [66.163.178.177]) by mx1.freebsd.org (Postfix) with SMTP id 5319A13C461 for ; Fri, 17 Aug 2007 01:28:52 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: (qmail 94556 invoked by uid 60001); 17 Aug 2007 01:02:10 -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=eYnLYzsRmlGVyi1TubLeaym6fMWdXhQc1VphHr1sCinpg/EQMk99we5NsMpxg9j56QXsg2t3l5/3nmOaYlrybKyUmJeBfRjIrYfpjvKViFPEZx1xiVVIA82hL+38ghVSRxSIAJnCclH1Tol0fhgT2v9X12Yxt3GqeJX58sl9/s4=; X-YMail-OSG: XMAsqP4VM1mmLFlS_o10VSLN0.OSknDPa4OktmS3IkPm23GlRycGisq59K156YfAcGG5ZXm0iQ-- Received: from [67.112.21.27] by web34511.mail.mud.yahoo.com via HTTP; Thu, 16 Aug 2007 18:02:09 PDT Date: Thu, 16 Aug 2007 18:02:09 -0700 (PDT) From: "N. Harrington" To: freebsd-fs@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <84373.93412.qm@web34511.mail.mud.yahoo.com> Subject: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 01:28:52 -0000 With ZFS, I have not seen much new going on with gjournal. I am curious what the status of gjournal and if it will likely be included with 6.3 (whenever that is due) Also, as of late, I have been using it with 6.2-STABLE via the patches and I seem to be getting 1/2 the transfer speeds compared to non journaled disks. It seems like this is recent as previous tests showed it as quite fast. Any suggestions on why this could be happening greatly appreciated. tested via dd if=/dev/zero of=./testfile bs=16 count=16384 On 2 different dual opteron systems with 8 gigs of Ram running FreeBSD 6.2-STABLE amd64 (as of 4 days ago) Regardless of the disk (I tested scsi and ata and one on a raid controller) both times I converted it to journal, the speed went in half. With disks getting larger and larger, why is it taking so long for a journaled filesystem to be standard on BSD? Thanks! Nicole From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 02:28:44 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 60ACB16A418 for ; Fri, 17 Aug 2007 02:28:44 +0000 (UTC) (envelope-from rick@kiwi-computer.com) Received: from kiwi-computer.com (keira.kiwi-computer.com [63.224.10.3]) by mx1.freebsd.org (Postfix) with SMTP id 2377713C45E for ; Fri, 17 Aug 2007 02:28:44 +0000 (UTC) (envelope-from rick@kiwi-computer.com) Received: (qmail 42750 invoked by uid 2001); 17 Aug 2007 02:02:02 -0000 Date: Thu, 16 Aug 2007 21:02:02 -0500 From: "Rick C. Petty" To: Gore Jarold Message-ID: <20070817020201.GA41414@keira.kiwi-computer.com> References: <723681.52797.qm@web63012.mail.re1.yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <723681.52797.qm@web63012.mail.re1.yahoo.com> User-Agent: Mutt/1.4.2.1i Cc: freebsd-fs@freebsd.org Subject: Re: vfs.ufs.dirhash_maxmem ... how high can I go ? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: rick-freebsd@kiwi-computer.com List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 02:28:44 -0000 On Wed, Aug 15, 2007 at 09:27:11AM -0700, Gore Jarold wrote: > > I have 4 GB of physical and 4 GB of swap, running on > normal 32-bit x86, and I have this set as well: Which means (on an x86 system) that you have 3 GB of physical RAM. If you don't believe me, look near the top of /var/run/dmesg.boot. -- Rick C. Petty From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 02:30:07 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 3980616A417 for ; Fri, 17 Aug 2007 02:30:07 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.ltdomains.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id 10E7213C45A for ; Fri, 17 Aug 2007 02:30:06 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from neutrino.vnode.org (r74-193-81-203.pfvlcmta01.grtntx.tl.dh.suddenlink.net [74.193.81.203]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l7H2U54Y008104 (version=TLSv1/SSLv3 cipher=DHE-DSS-AES256-SHA bits=256 verify=NO); Thu, 16 Aug 2007 21:30:06 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <46C50828.2020205@freebsd.org> Date: Thu, 16 Aug 2007 21:30:00 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.6 (X11/20070816) MIME-Version: 1.0 To: "N. Harrington" References: <84373.93412.qm@web34511.mail.mud.yahoo.com> In-Reply-To: <84373.93412.qm@web34511.mail.mud.yahoo.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-1.5 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: freebsd-fs@freebsd.org Subject: Re: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 02:30:07 -0000 On 08/16/07 20:02, N. Harrington wrote: > With ZFS, I have not seen much new going on with > gjournal. I am curious what the status of gjournal and > if it will likely be included with 6.3 (whenever that > is due) > > Also, as of late, I have been using it with > 6.2-STABLE via the patches and I seem to be getting > 1/2 the transfer speeds compared to non journaled > disks. It seems like this is recent as previous tests > showed it as quite fast. It would help if you included your gjournal configuration, file system configuration (how it was newfs'ed), what kind of performance you saw, and what you expected. > Any suggestions on why this could be happening > greatly appreciated. > > tested via > dd if=/dev/zero of=./testfile bs=16 count=16384 A 16byte block size is, well, really not going to prove much. Try a 1m block size. > On 2 different dual opteron systems with 8 gigs of > Ram running FreeBSD 6.2-STABLE amd64 (as of 4 days > ago) > > Regardless of the disk (I tested scsi and ata and one > on a raid controller) both times I converted it to > journal, the speed went in half. Half of what, down to what? > With disks getting larger and larger, why is it > taking so long for a journaled filesystem to be > standard on BSD? Since we've had soft-updates for quite some time, it has reduced the need for a journaled file system, however it is of course increasingly important. I suppose one has not happened yet because not enough developers have had the spare time to implement one. Besides, in 7-RELEASE, you'll have zfs, so there will be an option for you. You can always submit patches for journaling for one of the existing file systems if you'd like. Eric From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 12:29:12 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id E95B916A418 for ; Fri, 17 Aug 2007 12:29:12 +0000 (UTC) (envelope-from des@des.no) Received: from tim.des.no (tim.des.no [194.63.250.121]) by mx1.freebsd.org (Postfix) with ESMTP id A8AF013C461 for ; Fri, 17 Aug 2007 12:29:12 +0000 (UTC) (envelope-from des@des.no) Received: from tim.des.no (localhost [127.0.0.1]) by spam.des.no (Postfix) with ESMTP id 591AA20B1; Fri, 17 Aug 2007 14:29:08 +0200 (CEST) X-Spam-Tests: AWL X-Spam-Learn: disabled X-Spam-Score: 0.0/3.0 X-Spam-Checker-Version: SpamAssassin 3.2.1 (2007-05-02) on tim.des.no Received: from ds4.des.no (des.no [80.203.243.180]) by smtp.des.no (Postfix) with ESMTP id D4426208A; Fri, 17 Aug 2007 14:29:07 +0200 (CEST) Received: by ds4.des.no (Postfix, from userid 1001) id BFEDD8444F; Fri, 17 Aug 2007 14:29:07 +0200 (CEST) From: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= To: rick-freebsd@kiwi-computer.com References: <723681.52797.qm@web63012.mail.re1.yahoo.com> <20070817020201.GA41414@keira.kiwi-computer.com> Date: Fri, 17 Aug 2007 14:29:07 +0200 In-Reply-To: <20070817020201.GA41414@keira.kiwi-computer.com> (Rick C. Petty's message of "Thu\, 16 Aug 2007 21\:02\:02 -0500") Message-ID: <867inuibu4.fsf@ds4.des.no> User-Agent: Gnus/5.110006 (No Gnus v0.6) Emacs/22.1 (berkeley-unix) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Cc: freebsd-fs@freebsd.org, Gore Jarold Subject: Re: vfs.ufs.dirhash_maxmem ... how high can I go ? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 12:29:13 -0000 "Rick C. Petty" writes: > Gore Jarold writes: > > I have 4 GB of physical and 4 GB of swap, running on normal 32-bit > > x86, and I have this set as well: > Which means (on an x86 system) that you have 3 GB of physical RAM. No, it means has 4 GB of physical RAM, of which 3.5 GB are addressable. DES --=20 Dag-Erling Sm=C3=B8rgrav - des@des.no From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 15:42:40 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id DB0FA16A419 for ; Fri, 17 Aug 2007 15:42:40 +0000 (UTC) (envelope-from pjd@garage.freebsd.pl) Received: from mail.garage.freebsd.pl (arm132.internetdsl.tpnet.pl [83.17.198.132]) by mx1.freebsd.org (Postfix) with ESMTP id 7AC5513C469 for ; Fri, 17 Aug 2007 15:42:40 +0000 (UTC) (envelope-from pjd@garage.freebsd.pl) Received: by mail.garage.freebsd.pl (Postfix, from userid 65534) id F26D0487F0; Fri, 17 Aug 2007 17:42:37 +0200 (CEST) Received: from localhost (pjd.wheel.pl [10.0.1.1]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mail.garage.freebsd.pl (Postfix) with ESMTP id E821145683; Fri, 17 Aug 2007 17:42:32 +0200 (CEST) Date: Fri, 17 Aug 2007 17:41:39 +0200 From: Pawel Jakub Dawidek To: "N. Harrington" Message-ID: <20070817154139.GA1089@garage.freebsd.pl> References: <84373.93412.qm@web34511.mail.mud.yahoo.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="GvXjxJ+pjyke8COw" Content-Disposition: inline In-Reply-To: <84373.93412.qm@web34511.mail.mud.yahoo.com> User-Agent: Mutt/1.4.2.3i X-PGP-Key-URL: http://people.freebsd.org/~pjd/pjd.asc X-OS: FreeBSD 7.0-CURRENT i386 X-Spam-Checker-Version: SpamAssassin 3.0.4 (2005-06-05) on mail.garage.freebsd.pl X-Spam-Level: X-Spam-Status: No, score=-5.9 required=3.0 tests=ALL_TRUSTED,BAYES_00 autolearn=ham version=3.0.4 Cc: freebsd-fs@freebsd.org Subject: Re: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 15:42:40 -0000 --GvXjxJ+pjyke8COw Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Aug 16, 2007 at 06:02:09PM -0700, N. Harrington wrote: > With ZFS, I have not seen much new going on with > gjournal. I am curious what the status of gjournal and > if it will likely be included with 6.3 (whenever that > is due) That was the plan, but I'm not yet sure if I'll find time to do it. > Also, as of late, I have been using it with > 6.2-STABLE via the patches and I seem to be getting > 1/2 the transfer speeds compared to non journaled > disks. It seems like this is recent as previous tests > showed it as quite fast. >=20 > Any suggestions on why this could be happening > greatly appreciated. >=20 > tested via=20 > dd if=3D/dev/zero of=3D./testfile bs=3D16 count=3D16384 Gjournal is ~two times faster than UFS+SU for many small, random and parallel writes (such as running few 'tar x' processes), but is two times slower than UFS+SU for one sequential write stream, as there is no much that can be optimized there. > With disks getting larger and larger, why is it > taking so long for a journaled filesystem to be > standard on BSD? We have ZFS now, we don't need journaled file system:) --=20 Pawel Jakub Dawidek http://www.wheel.pl pjd@FreeBSD.org http://www.FreeBSD.org FreeBSD committer Am I Evil? Yes, I Am! --GvXjxJ+pjyke8COw Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFGxcGzForvXbEpPzQRAnX4AKCBQVAAhqJ5A/FUME2Zw+y++mlbFQCg+Xwf RKFNGrKlhENuO0DIBbMx9rk= =UuH1 -----END PGP SIGNATURE----- --GvXjxJ+pjyke8COw-- From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 16:18:30 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 58A3B16A418 for ; Fri, 17 Aug 2007 16:18:30 +0000 (UTC) (envelope-from cracauer@koef.zs64.net) Received: from koef.zs64.net (koef.zs64.net [212.12.50.230]) by mx1.freebsd.org (Postfix) with ESMTP id EAD4013C48E for ; Fri, 17 Aug 2007 16:18:29 +0000 (UTC) (envelope-from cracauer@koef.zs64.net) Received: from koef.zs64.net (koef.zs64.net [212.12.50.230]) by koef.zs64.net (8.14.1/8.14.1) with ESMTP id l7HFvrFE007717 for ; Fri, 17 Aug 2007 17:57:53 +0200 (CEST) (envelope-from cracauer@koef.zs64.net) Received: (from cracauer@localhost) by koef.zs64.net (8.14.1/8.14.1/Submit) id l7HFvqeS007716 for freebsd-fs@freebsd.org; Fri, 17 Aug 2007 11:57:52 -0400 (EDT) (envelope-from cracauer) Date: Fri, 17 Aug 2007 11:57:52 -0400 From: Martin Cracauer To: freebsd-fs@freebsd.org Message-ID: <20070817155741.GA6255@cons.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4.2.2i Subject: Block device over network from Linux to FreeBSD X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 16:18:30 -0000 All right, here's a question that'll make your IQ drop by 5 points just from pondering it :-) What's the best way to provide, over the network, a block device on harddrives that live on a Linux box and "export" them to a FreeBSD machine? Aka I want a FreeBSD filesystem on harddrives that are physically in a Linux box. Long story: My backup strategy is a FreeBSD filesystem with snapshots on a bunch of harddrives that live on networked computers in the basement. All these computers boot diskless or disky into a variety of OSes, usually Linux or FreeBSD. It would be easy to just use ext2fs or another filesystem supported by both, but I'd really like ufs2 snapshots. So I need to access the disks in a box running Linux as a block device from a machine running FreeBSD. When the machine having the physical disks runs FreeBSD I want to access the same raw devices directly, of course. The brute-force approach would be: - ext2fs on disks - files inside ext2fs for use via mdconfig (and ccd) - then, depending on OSes booted, either: - export via NFS and mdconfig on NFS mounts on remote FreeBSD machine - direct FreeBSD mount (machine runs FreeBSD) Another alternative I see is VMware or if any of the free emulators can boot FreeBSD on Linux and use the disks directly in the guest OS. Linux has a network layer for block devices: http://www.it.uc3m.es/ptb/nbd/ . On first sight, it doesn't look too exiting nor does it look straightforward to implement a client in GEOM. It uses daemons on both ends, so failover will not exactly be an improvement over NFS. At least with NFS you know that a lot of other people depend on what you write being delivered eventually. Then there's ATA over Ethernet as an established protocol. Any other ideas? USB'ing the harddrives is not considered sportish :-) Martin -- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Martin Cracauer http://www.cons.org/cracauer/ FreeBSD - where you want to go, today. http://www.freebsd.org/ From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 17:36:14 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2558E16A419 for ; Fri, 17 Aug 2007 17:36:14 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.ltdomains.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id ED80413C480 for ; Fri, 17 Aug 2007 17:36:13 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from proton.local (209-163-168-124.static.twtelecom.net [209.163.168.124]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l7HHaAIQ072732; Fri, 17 Aug 2007 12:36:10 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <46C5DC8C.50106@freebsd.org> Date: Fri, 17 Aug 2007 12:36:12 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.6 (Macintosh/20070728) MIME-Version: 1.0 To: Martin Cracauer References: <20070817155741.GA6255@cons.org> In-Reply-To: <20070817155741.GA6255@cons.org> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-1.6 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: freebsd-fs@freebsd.org Subject: Re: Block device over network from Linux to FreeBSD X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 17:36:14 -0000 Martin Cracauer wrote: > All right, here's a question that'll make your IQ drop by 5 points > just from pondering it :-) > > What's the best way to provide, over the network, a block device on > harddrives that live on a Linux box and "export" them to a FreeBSD > machine? Aka I want a FreeBSD filesystem on harddrives that are > physically in a Linux box. > > Long story: > > My backup strategy is a FreeBSD filesystem with snapshots on a bunch > of harddrives that live on networked computers in the basement. All > these computers boot diskless or disky into a variety of OSes, usually > Linux or FreeBSD. It would be easy to just use ext2fs or another > filesystem supported by both, but I'd really like ufs2 snapshots. So > I need to access the disks in a box running Linux as a block device > from a machine running FreeBSD. When the machine having the physical > disks runs FreeBSD I want to access the same raw devices directly, of > course. > > The brute-force approach would be: > - ext2fs on disks > - files inside ext2fs for use via mdconfig (and ccd) > - then, depending on OSes booted, either: > - export via NFS and mdconfig on NFS mounts on remote FreeBSD machine > - direct FreeBSD mount (machine runs FreeBSD) > > Another alternative I see is VMware or if any of the free emulators > can boot FreeBSD on Linux and use the disks directly in the guest OS. > > Linux has a network layer for block devices: > http://www.it.uc3m.es/ptb/nbd/ . On first sight, it doesn't look too > exiting nor does it look straightforward to implement a client in > GEOM. It uses daemons on both ends, so failover will not exactly be > an improvement over NFS. At least with NFS you know that a lot of > other people depend on what you write being delivered eventually. > > Then there's ATA over Ethernet as an established protocol. > > Any other ideas? > > USB'ing the harddrives is not considered sportish :-) > > Martin iSCSI? Eric From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 18:51:01 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D64B016A421 for ; Fri, 17 Aug 2007 18:51:01 +0000 (UTC) (envelope-from freebsd-fs@m.gmane.org) Received: from ciao.gmane.org (main.gmane.org [80.91.229.2]) by mx1.freebsd.org (Postfix) with ESMTP id 8E5BC13C4A3 for ; Fri, 17 Aug 2007 18:51:01 +0000 (UTC) (envelope-from freebsd-fs@m.gmane.org) Received: from list by ciao.gmane.org with local (Exim 4.43) id 1IM6uI-0006tp-7u for freebsd-fs@freebsd.org; Fri, 17 Aug 2007 20:50:50 +0200 Received: from 78-0-71-152.adsl.net.t-com.hr ([78.0.71.152]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Fri, 17 Aug 2007 20:50:50 +0200 Received: from ivoras by 78-0-71-152.adsl.net.t-com.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Fri, 17 Aug 2007 20:50:50 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-fs@freebsd.org From: Ivan Voras Date: Fri, 17 Aug 2007 20:50:26 +0200 Lines: 32 Message-ID: References: <20070817155741.GA6255@cons.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enigF60D69608859BDF93FEF092B" X-Complaints-To: usenet@sea.gmane.org X-Gmane-NNTP-Posting-Host: 78-0-71-152.adsl.net.t-com.hr User-Agent: Thunderbird 1.5.0.12 (Windows/20070509) In-Reply-To: <20070817155741.GA6255@cons.org> X-Enigmail-Version: 0.94.3.0 Sender: news Subject: Re: Block device over network from Linux to FreeBSD X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 18:51:01 -0000 This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigF60D69608859BDF93FEF092B Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Martin Cracauer wrote: > All right, here's a question that'll make your IQ drop by 5 points > just from pondering it :-) > Any other ideas? If you're into programming, you can port the ggatec daemon that serves as an end-point for ggated. Ggatec is userland-only and it just does IO to/from a local file or device, nothing fancy (something like completely nonstandard variant of iSCSI, ATA over ethernet, etc.). --------------enigF60D69608859BDF93FEF092B Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.5 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGxe33ldnAQVacBcgRAn2nAKCxr9//i5YCzY3utewbn9iCT+oGrQCbBNn6 YfajrxWNAGiD/XYFeMJPzI0= =4UnH -----END PGP SIGNATURE----- --------------enigF60D69608859BDF93FEF092B-- From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 20:03:07 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2EB2616A419 for ; Fri, 17 Aug 2007 20:03:07 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: from web34506.mail.mud.yahoo.com (web34506.mail.mud.yahoo.com [66.163.178.172]) by mx1.freebsd.org (Postfix) with SMTP id E9AA113C480 for ; Fri, 17 Aug 2007 20:03:06 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: (qmail 63621 invoked by uid 60001); 17 Aug 2007 20:03:06 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=OSaRQ91HSqfsVIfBvX70c3wNQ71zwN3kcKvY7utJCn30eSecsHVC4gjv5Lsb+21JCNX+LEvSijYdNDyhYbjoywJnbweQgDMXTSAF1bwp7ikwEAARH1IjQki/b/U7ofbpZZl0KgBSQTTW7lg5J8SXEmbZ90DkWpZCrWhcmHEBXVA=; X-YMail-OSG: YJ_C9yYVM1mqwR3LYoSheRq7zPD1.7GtLU1u7IXerPiU25kYUAUVDQSPT_QjeraeUf0rI6e_OGBWJzdlyzimPkHKlBrBotrflHIOar_DFJ66.yL6OBk- Received: from [67.112.21.27] by web34506.mail.mud.yahoo.com via HTTP; Fri, 17 Aug 2007 13:03:06 PDT Date: Fri, 17 Aug 2007 13:03:06 -0700 (PDT) From: "N. Harrington" To: Pawel Jakub Dawidek In-Reply-To: <20070817154139.GA1089@garage.freebsd.pl> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <236419.62551.qm@web34506.mail.mud.yahoo.com> Cc: freebsd-fs@freebsd.org Subject: Re: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 20:03:07 -0000 --- Pawel Jakub Dawidek wrote: > On Thu, Aug 16, 2007 at 06:02:09PM -0700, N. > Harrington wrote: > > With ZFS, I have not seen much new going on with > > gjournal. I am curious what the status of gjournal > and > > if it will likely be included with 6.3 (whenever > that > > is due) > > That was the plan, but I'm not yet sure if I'll find > time to do it. I hope so. It seems like everything is there. > > Also, as of late, I have been using it with > > 6.2-STABLE via the patches and I seem to be > getting > > 1/2 the transfer speeds compared to non journaled > > disks. It seems like this is recent as previous > tests > > showed it as quite fast. > > > > Any suggestions on why this could be happening > > greatly appreciated. > > > > tested via > > dd if=/dev/zero of=./testfile bs=16 count=16384 > > Gjournal is ~two times faster than UFS+SU for many > small, random and > parallel writes (such as running few 'tar x' > processes), but is two > times slower than UFS+SU for one sequential write > stream, as there is no > much that can be optimized there. Is there a test that I could try? I would swear that using the same test in the past, it showed as faster. > > With disks getting larger and larger, why is it > > taking so long for a journaled filesystem to be > > standard on BSD? > > We have ZFS now, we don't need journaled file > system:) I guess I need to learn now about ZFS. As far as I know it is not available in 6 yet and it is for raid, not for journaling or a filesystem itself. It would seem silly to have to resort to using zfs just so I can have a disk/mount that won't not require days to fsck (9 times out of 10) if there should be a problem. gjournaling has been the only thing that has kept FreeBSD used at my company for file storage. I have many many active TB's with it. Much of it in 500G slices, which I was thinking of increasing to 1TB slices. I could never do that with UFS with any sanity or hope of having a job after the day long fsck. I certainly could never justify why we should not switch to Linux to prevent it otherwise. I am so grateful you did gjournaling. If I thought we could afford it, we would pay to make sure its included in 6.3. Nicole > -- > Pawel Jakub Dawidek > http://www.wheel.pl > pjd@FreeBSD.org > http://www.FreeBSD.org > FreeBSD committer Am I Evil? > Yes, I Am! > From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 20:27:52 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 3ACD716A418 for ; Fri, 17 Aug 2007 20:27:52 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: from web34513.mail.mud.yahoo.com (web34513.mail.mud.yahoo.com [66.163.178.179]) by mx1.freebsd.org (Postfix) with SMTP id 0CD0F13C428 for ; Fri, 17 Aug 2007 20:27:51 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: (qmail 69059 invoked by uid 60001); 17 Aug 2007 20:27:51 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=y0ews2jBVVjAB+5VKkh0ErWdWp0Azg2fibtyB1SRVCgDj0wUEw7ybS3BCG+KfpM1QKS1AHIn9Ld8wNZo4ROjQuCN2FlSxitLj7xjCnczzKsYf0Dz9vHi1LGiTEyJ3etBW1cPpb6E51vNlE58lnRoAFMIXvsaf7+dPLvdShESM8g=; X-YMail-OSG: S60k.4kVM1lAJwPo3PgOR4PwKwxsKUns02dACdUEdkmm2hVhT7URWMFTpuG45bZyRGR4utAZWw-- Received: from [67.112.21.27] by web34513.mail.mud.yahoo.com via HTTP; Fri, 17 Aug 2007 13:27:50 PDT Date: Fri, 17 Aug 2007 13:27:50 -0700 (PDT) From: "N. Harrington" To: Eric Anderson , freebsd-fs@freebsd.org In-Reply-To: <46C50828.2020205@freebsd.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <276388.68929.qm@web34513.mail.mud.yahoo.com> Cc: Subject: Re: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 20:27:52 -0000 --- Eric Anderson wrote: > On 08/16/07 20:02, N. Harrington wrote: > > With ZFS, I have not seen much new going on with > > gjournal. I am curious what the status of gjournal > and > > if it will likely be included with 6.3 (whenever > that > > is due) > > > > Also, as of late, I have been using it with > > 6.2-STABLE via the patches and I seem to be > getting > > 1/2 the transfer speeds compared to non journaled > > disks. It seems like this is recent as previous > tests > > showed it as quite fast. > > It would help if you included your gjournal > configuration, file system > configuration (how it was newfs'ed), what kind of > performance you saw, > and what you expected. I left it "vague" as I was trying it on 3 different setups. Also, since whatever the speed was with UFS, became 1/2 with journal. So were I started from, each being different, does not matter or really help. I did a standard newfs ( newfs /dev/da1s1e and newfs /dev/da1s1d ) before the journaling. After I edited the bsdlabel to have the 16 offset. I have seen that there is a newfs -J, however I am uncertain of its usage (or what it does) if one is using a separate (beginning) slice for the journal. Per Pawel, it seems perhaps my test is flawed. However I would swear that doing this test before, showed an improvement, not decrease. Hence I wonder if there has been some recent change in 6-STABLE such that it does not integrate as well somehow. > > > Any suggestions on why this could be happening > > greatly appreciated. > > > > tested via > > dd if=/dev/zero of=./testfile bs=16 count=16384 > > A 16byte block size is, well, really not going to > prove much. Try a 1m > block size. If I were using larger files, perhaps. But I store many smaller files, so this seems much more realistic. I will perhaps try some slightly larger sizes. > > On 2 different dual opteron systems with 8 gigs > of > > Ram running FreeBSD 6.2-STABLE amd64 (as of 4 days > > ago) > > > > Regardless of the disk (I tested scsi and ata and > one > > on a raid controller) both times I converted it to > > journal, the speed went in half. > > Half of what, down to what? Why does that matter? If its 1/2, why does it matter what it is 1/2 of? > > With disks getting larger and larger, why is it > > taking so long for a journaled filesystem to be > > standard on BSD? > > > Since we've had soft-updates for quite some time, it > has reduced the > need for a journaled file system, however it is of > course increasingly > important. I suppose one has not happened yet > because not enough > developers have had the spare time to implement one. > Besides, in > 7-RELEASE, you'll have zfs, so there will be an > option for you. You can > always submit patches for journaling for one of the > existing file > systems if you'd like. That assumes that I am a developer with such skills. People always saying, then submit a patch, assumes one has the skills and time to do so. I, and most companies, would rather pay to have someone who has the specialized skills to do the job well. I, and my company contribute money to BSD, as that is the resource we have available. I have seem some people solicit money requests to do enhancement work etc.. Usually with good results. It seems to me, more would get done that way than always assuming anyone who knows how to use freeBSD well and or post on the mailing list, has the ability to also write code for it. (my tiny rant) Be well. Nicole > Eric > > > > > > > > From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 22:51:05 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2905016A419 for ; Fri, 17 Aug 2007 22:51:05 +0000 (UTC) (envelope-from rick@kiwi-computer.com) Received: from kiwi-computer.com (keira.kiwi-computer.com [63.224.10.3]) by mx1.freebsd.org (Postfix) with SMTP id A89EE13C46E for ; Fri, 17 Aug 2007 22:51:04 +0000 (UTC) (envelope-from rick@kiwi-computer.com) Received: (qmail 92677 invoked by uid 2001); 17 Aug 2007 22:51:03 -0000 Date: Fri, 17 Aug 2007 17:51:03 -0500 From: "Rick C. Petty" To: Dag-Erling Sm??rgrav Message-ID: <20070817225103.GA92437@keira.kiwi-computer.com> References: <723681.52797.qm@web63012.mail.re1.yahoo.com> <20070817020201.GA41414@keira.kiwi-computer.com> <867inuibu4.fsf@ds4.des.no> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <867inuibu4.fsf@ds4.des.no> User-Agent: Mutt/1.4.2.1i Cc: freebsd-fs@freebsd.org, Gore Jarold Subject: Re: vfs.ufs.dirhash_maxmem ... how high can I go ? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: rick-freebsd@kiwi-computer.com List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 22:51:05 -0000 On Fri, Aug 17, 2007 at 02:29:07PM +0200, Dag-Erling Sm??rgrav wrote: > "Rick C. Petty" writes: > > Gore Jarold writes: > > > I have 4 GB of physical and 4 GB of swap, running on normal 32-bit > > > x86, and I have this set as well: > > Which means (on an x86 system) that you have 3 GB of physical RAM. > > No, it means has 4 GB of physical RAM, of which 3.5 GB are addressable. I've never seen FreeBSD address more than 3.0 GB of RAM without PAE.. it always seems to reserve 1.0 GB for video & other mmap'd I/O. I've tried this in systems with 4 MB video cards and most devices disabled. I thought you had to tweak some sysctl in freebsd to force the kernel not to map that last 1 GB. -- Rick C. Petty From owner-freebsd-fs@FreeBSD.ORG Fri Aug 17 22:52:42 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8A7CC16A418 for ; Fri, 17 Aug 2007 22:52:42 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: from web34515.mail.mud.yahoo.com (web34515.mail.mud.yahoo.com [66.163.178.181]) by mx1.freebsd.org (Postfix) with SMTP id 5C59C13C49D for ; Fri, 17 Aug 2007 22:52:42 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: (qmail 26268 invoked by uid 60001); 17 Aug 2007 22:52:41 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=HXaOpJKPZvm6SAXm9IbKYKstHQtfPIPmXw05p1t/61Wl7RVSqVzoS/aCDWYJcz/kdomeTZl0G2Ai8BRCHxCZsH7dTOMoi04JdmkMHJUp+ZUqf2YtaJ1xOEOFWB3EQ7GpvkDbUUBw9APQA35ZeG1BzQVeeCGZAjna4nHIvQsrdmk=; X-YMail-OSG: BCaX9QAVM1kEf69AEwumSf5DbBGJDmlhXloJ.MJ.KlvXO3fpxhxwjU1S_EJeOXH_bnpff4unRw-- Received: from [67.112.21.27] by web34515.mail.mud.yahoo.com via HTTP; Fri, 17 Aug 2007 15:52:41 PDT Date: Fri, 17 Aug 2007 15:52:41 -0700 (PDT) From: "N. Harrington" To: Pawel Jakub Dawidek In-Reply-To: <20070817154139.GA1089@garage.freebsd.pl> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <718389.22856.qm@web34515.mail.mud.yahoo.com> Cc: freebsd-fs@freebsd.org Subject: Re: Gjournal reporting 1/2 the speed of non journaled? - examples X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 17 Aug 2007 22:52:42 -0000 --- Pawel Jakub Dawidek wrote: > On Thu, Aug 16, 2007 at 06:02:09PM -0700, N. > Harrington wrote: > > With ZFS, I have not seen much new going on with > > gjournal. I am curious what the status of gjournal > and > > if it will likely be included with 6.3 (whenever > that > > is due) > > That was the plan, but I'm not yet sure if I'll find > time to do it. > > > Also, as of late, I have been using it with > > 6.2-STABLE via the patches and I seem to be > getting > > 1/2 the transfer speeds compared to non journaled > > disks. It seems like this is recent as previous > tests > > showed it as quite fast. > > > > Any suggestions on why this could be happening > > greatly appreciated. > > > > tested via > > dd if=/dev/zero of=./testfile bs=16 count=16384 > > Gjournal is ~two times faster than UFS+SU for many > small, random and > parallel writes (such as running few 'tar x' > processes), but is two > times slower than UFS+SU for one sequential write > stream, as there is no > much that can be optimized there. > > > With disks getting larger and larger, why is it > > taking so long for a journaled filesystem to be > > standard on BSD? > > We have ZFS now, we don't need journaled file > system:) Here is an example of my tests and what I am experiencing. FreeBSD server.old 6.2-PRERELEASE FreeBSD 6.2-PRERELEASE #1: Wed Nov 22 21:20:08 PST 2006 With Gjournal: root@:/1-journal> time dd if=/dev/zero of=/1-journal/testfile1 bs=16k count=16384 16384+0 records in 16384+0 records out 268435456 bytes transferred in 10.257279 secs (26170240 bytes/sec) dd if=/dev/zero of=/cache1/testfile1 bs=16k count=16384 0.02s user 0.80s system 7% cpu 10.393 total Without Gjournal: root@1:/2-UFS> time dd if=/dev/zero of=/2/testfile1 bs=16k count=16384 16384+0 records in 16384+0 records out 268435456 bytes transferred in 26.682630 secs (10060307 bytes/sec) dd if=/dev/zero of=/home/testfile1 bs=16k count=16384 0.01s user 0.61s system 2% cpu 28.448 total As you can see, with this test, using Gjournaling tested faster. Now on a new build: FreeBSD newtest 6.2-STABLE FreeBSD 6.2-STABLE #1: Tue Aug 14 14:31:53 PDT 2007 nicole@newtest:/usr/obj/spare/src/sys/GENERAL amd64 Without Gjournal root@newtest:/home/nicole> time dd if=/dev/zero of=./testfile bs=16k count=16384 16384+0 records in 16384+0 records out 268435456 bytes transferred in 4.749401 secs (56519854 bytes/sec) dd if=/dev/zero of=./testfile bs=16k count=16384 0.01s user 0.45s system 9% cpu 4.751 total With Gjournaling: root@newtest:/home/nicole> time dd if=/dev/zero of=/tmp/testfile bs=16k count=16384 16384+0 records in 16384+0 records out 268435456 bytes transferred in 7.505348 secs (35765891 bytes/sec) dd if=/dev/zero of=/tmp/testfile bs=16k count=16384 0.00s user 0.57s system 7% cpu 7.517 total So as shown here, my test on an older 6.2 (with same patches) showed improved performance. But now on a new 6.2 build system, shows decreased performance. Nicole From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 00:10:07 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2E90616A474 for ; Sat, 18 Aug 2007 00:10:07 +0000 (UTC) (envelope-from freebsd-fs@m.gmane.org) Received: from ciao.gmane.org (main.gmane.org [80.91.229.2]) by mx1.freebsd.org (Postfix) with ESMTP id CD13A13C4E7 for ; Sat, 18 Aug 2007 00:10:01 +0000 (UTC) (envelope-from freebsd-fs@m.gmane.org) Received: from list by ciao.gmane.org with local (Exim 4.43) id 1IMBt6-0004HR-7a for freebsd-fs@freebsd.org; Sat, 18 Aug 2007 02:09:56 +0200 Received: from 78-0-71-152.adsl.net.t-com.hr ([78.0.71.152]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 18 Aug 2007 02:09:56 +0200 Received: from ivoras by 78-0-71-152.adsl.net.t-com.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 18 Aug 2007 02:09:56 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-fs@freebsd.org From: Ivan Voras Date: Sat, 18 Aug 2007 02:09:29 +0200 Lines: 30 Message-ID: References: <84373.93412.qm@web34511.mail.mud.yahoo.com> <20070817154139.GA1089@garage.freebsd.pl> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enig6F643DFB9343F70128D39D49" X-Complaints-To: usenet@sea.gmane.org X-Gmane-NNTP-Posting-Host: 78-0-71-152.adsl.net.t-com.hr User-Agent: Thunderbird 2.0.0.6 (Windows/20070728) In-Reply-To: <20070817154139.GA1089@garage.freebsd.pl> X-Enigmail-Version: 0.95.3 Sender: news Subject: Re: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 00:10:07 -0000 This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig6F643DFB9343F70128D39D49 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Pawel Jakub Dawidek wrote: > We have ZFS now, we don't need journaled file system:) I hope development on other file systems on FreeBSD won't stop because of ZFS :) Choice is always good and some file systems are better in some tasks than the other, and compatibility with other OSes is always good (for one, I'd love ext3 on FreeBSD :) ) --------------enig6F643DFB9343F70128D39D49 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.5 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGxjjDldnAQVacBcgRAqXaAJ4i0JHKp+DCowZ0a2f+2aZaQLx1qACeN4DR 2hguUdWpFwhTxIZ871E7Iw8= =XC73 -----END PGP SIGNATURE----- --------------enig6F643DFB9343F70128D39D49-- From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 04:18:20 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id C3A3616A420 for ; Sat, 18 Aug 2007 04:18:20 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.ltdomains.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id A290A13C461 for ; Sat, 18 Aug 2007 04:18:20 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from proton.local (r74-193-81-203.pfvlcmta01.grtntx.tl.dh.suddenlink.net [74.193.81.203]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l7I4IHaG055360; Fri, 17 Aug 2007 23:18:18 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <46C67305.3080308@freebsd.org> Date: Fri, 17 Aug 2007 23:18:13 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.6 (Macintosh/20070728) MIME-Version: 1.0 To: "N. Harrington" References: <276388.68929.qm@web34513.mail.mud.yahoo.com> In-Reply-To: <276388.68929.qm@web34513.mail.mud.yahoo.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-1.6 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: freebsd-fs@freebsd.org Subject: Re: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 04:18:21 -0000 N. Harrington wrote: > --- Eric Anderson wrote: > >> On 08/16/07 20:02, N. Harrington wrote: >>> With ZFS, I have not seen much new going on with >>> gjournal. I am curious what the status of gjournal >> and >>> if it will likely be included with 6.3 (whenever >> that >>> is due) >>> >>> Also, as of late, I have been using it with >>> 6.2-STABLE via the patches and I seem to be >> getting >>> 1/2 the transfer speeds compared to non journaled >>> disks. It seems like this is recent as previous >> tests >>> showed it as quite fast. >> It would help if you included your gjournal >> configuration, file system >> configuration (how it was newfs'ed), what kind of >> performance you saw, >> and what you expected. > > I left it "vague" as I was trying it on 3 different > setups. Also, since whatever the speed was with UFS, > became 1/2 with journal. So were I started from, each > being different, does not matter or really help. > > I did a standard newfs ( newfs /dev/da1s1e and newfs > /dev/da1s1d ) before the journaling. After I edited > the bsdlabel to have the 16 offset. > > I have seen that there is a newfs -J, however I am > uncertain of its usage (or what it does) if one is > using a separate (beginning) slice for the journal. > > Per Pawel, it seems perhaps my test is flawed. However > I would swear that doing this test before, showed an > improvement, not decrease. Hence I wonder if there has > been some recent change in 6-STABLE such that it does > not integrate as well somehow. The comments above are exactly why the details on configuration are important. You'll find better performance if you can put the journal on a different disk than the data. >>> Any suggestions on why this could be happening >>> greatly appreciated. >>> >>> tested via >>> dd if=/dev/zero of=./testfile bs=16 count=16384 >> A 16byte block size is, well, really not going to >> prove much. Try a 1m >> block size. > > If I were using larger files, perhaps. But I store > many smaller files, so this seems much more realistic. > I will perhaps try some slightly larger sizes. Which is why your test is not indicative of your real use. You are sending sequential writes to the disk, when you really are using it in a more random fashion. You should probably use iozone, or craft a simple loop around a bunch of dd commands that do single files each, or something. >>> On 2 different dual opteron systems with 8 gigs >> of >>> Ram running FreeBSD 6.2-STABLE amd64 (as of 4 days >>> ago) >>> >>> Regardless of the disk (I tested scsi and ata and >> one >>> on a raid controller) both times I converted it to >>> journal, the speed went in half. >> Half of what, down to what? > > Why does that matter? If its 1/2, why does it matter > what it is 1/2 of? It's another data point. If you want help debugging something, it doesn't hurt to include extra information, whereas arguing about whether it's important or not is kind of a waste of time.. :( >>> With disks getting larger and larger, why is it >>> taking so long for a journaled filesystem to be >>> standard on BSD? >> >> Since we've had soft-updates for quite some time, it >> has reduced the >> need for a journaled file system, however it is of >> course increasingly >> important. I suppose one has not happened yet >> because not enough >> developers have had the spare time to implement one. >> Besides, in >> 7-RELEASE, you'll have zfs, so there will be an >> option for you. You can >> always submit patches for journaling for one of the >> existing file >> systems if you'd like. > > That assumes that I am a developer with such skills. > People always saying, then submit a patch, assumes one > has the skills and time to do so. I, and most > companies, would rather pay to have someone who has > the specialized skills to do the job well. I, and my > company contribute money to BSD, as that is the > resource we have available. I have seem some people > solicit money requests to do enhancement work etc.. > Usually with good results. It seems to me, more would > get done that way than always assuming anyone who > knows how to use freeBSD well and or post on the > mailing list, has the ability to also write code for > it. > (my tiny rant) Understood. I wasn't trying to say 'do it yourself', I don't know your skill level. I'm just saying if someone wants it, they should contribute to the effort - be it code, support, money, advocacy, etc. Sounds like you are doing well to contribute with your company. What I recommend, is if it is important to you, get your company to put up a bounty for what you want. Other companies/individuals may also chip in and add to your bounty. Look at the rsync.net site for a great example. Eric From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 04:22:08 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 0CA8116A419; Sat, 18 Aug 2007 04:22:08 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.ltdomains.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id DD93813C45A; Sat, 18 Aug 2007 04:22:07 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from proton.local (r74-193-81-203.pfvlcmta01.grtntx.tl.dh.suddenlink.net [74.193.81.203]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l7I4M6QO055469; Fri, 17 Aug 2007 23:22:06 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <46C673EA.6010106@freebsd.org> Date: Fri, 17 Aug 2007 23:22:02 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.6 (Macintosh/20070728) MIME-Version: 1.0 To: "N. Harrington" References: <236419.62551.qm@web34506.mail.mud.yahoo.com> In-Reply-To: <236419.62551.qm@web34506.mail.mud.yahoo.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-1.6 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: freebsd-fs@freebsd.org, Pawel Jakub Dawidek Subject: Re: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 04:22:08 -0000 N. Harrington wrote: > --- Pawel Jakub Dawidek wrote: > >> On Thu, Aug 16, 2007 at 06:02:09PM -0700, N. >> Harrington wrote: >>> With ZFS, I have not seen much new going on with >>> gjournal. I am curious what the status of gjournal >> and >>> if it will likely be included with 6.3 (whenever >> that >>> is due) >> That was the plan, but I'm not yet sure if I'll find >> time to do it. > > I hope so. It seems like everything is there. At my previous employer, we used it heavily, and it worked well. However, I do think that there are some patches that have gone into 7.0 that aren't in 6 with regards to locking/vfs/etc, that remove some bugs. I saw deadlocks once in a while on 6, but after moving to 7, they go away. >>> Also, as of late, I have been using it with >>> 6.2-STABLE via the patches and I seem to be >> getting >>> 1/2 the transfer speeds compared to non journaled >>> disks. It seems like this is recent as previous >> tests >>> showed it as quite fast. >>> >>> Any suggestions on why this could be happening >>> greatly appreciated. >>> >>> tested via >>> dd if=/dev/zero of=./testfile bs=16 count=16384 >> Gjournal is ~two times faster than UFS+SU for many >> small, random and >> parallel writes (such as running few 'tar x' >> processes), but is two >> times slower than UFS+SU for one sequential write >> stream, as there is no >> much that can be optimized there. > > Is there a test that I could try? I would swear that > using the same test in the past, it showed as faster. > > >>> With disks getting larger and larger, why is it >>> taking so long for a journaled filesystem to be >>> standard on BSD? >> We have ZFS now, we don't need journaled file >> system:) > > I guess I need to learn now about ZFS. As far as I > know it is not available in 6 yet and it is for raid, > not for journaling or a filesystem itself. ZFS is a file system, with a lot of 'RAID'-like storage stuff built in. You should read about it a bit - very cool stuff. It's incredibly impressive what Pawel did in the amount of time he did it. > It would seem silly to have to resort to using zfs > just so I can have a disk/mount that won't not require > days to fsck (9 times out of 10) if there should be a > problem. gjournaling has been the only thing that has > kept FreeBSD used at my company for file storage. I > have many many active TB's with it. Much of it in 500G > slices, which I was thinking of increasing to 1TB > slices. I could never do that with UFS with any sanity > or hope of having a job after the day long fsck. I > certainly could never justify why we should not switch > to Linux to prevent it otherwise. Unfortunately, this is very very true. > I am so grateful you did gjournaling. If I thought we > could afford it, we would pay to make sure its > included in 6.3. > > > Nicole > > >> -- >> Pawel Jakub Dawidek >> http://www.wheel.pl >> pjd@FreeBSD.org >> http://www.FreeBSD.org >> FreeBSD committer Am I Evil? >> Yes, I Am! >> > > _______________________________________________ > freebsd-fs@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-fs > To unsubscribe, send any mail to "freebsd-fs-unsubscribe@freebsd.org" From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 04:30:33 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 8422F16A417; Sat, 18 Aug 2007 04:30:33 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.ltdomains.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id 607D813C48E; Sat, 18 Aug 2007 04:30:33 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from proton.local (r74-193-81-203.pfvlcmta01.grtntx.tl.dh.suddenlink.net [74.193.81.203]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l7I4UWLO055686; Fri, 17 Aug 2007 23:30:32 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <46C675E3.1000001@freebsd.org> Date: Fri, 17 Aug 2007 23:30:27 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.6 (Macintosh/20070728) MIME-Version: 1.0 To: "N. Harrington" References: <718389.22856.qm@web34515.mail.mud.yahoo.com> In-Reply-To: <718389.22856.qm@web34515.mail.mud.yahoo.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-1.7 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: freebsd-fs@freebsd.org, Pawel Jakub Dawidek Subject: Re: Gjournal reporting 1/2 the speed of non journaled? - examples X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 04:30:33 -0000 N. Harrington wrote: > --- Pawel Jakub Dawidek wrote: > >> On Thu, Aug 16, 2007 at 06:02:09PM -0700, N. >> Harrington wrote: >>> With ZFS, I have not seen much new going on with >>> gjournal. I am curious what the status of gjournal >> and >>> if it will likely be included with 6.3 (whenever >> that >>> is due) >> That was the plan, but I'm not yet sure if I'll find >> time to do it. >> >>> Also, as of late, I have been using it with >>> 6.2-STABLE via the patches and I seem to be >> getting >>> 1/2 the transfer speeds compared to non journaled >>> disks. It seems like this is recent as previous >> tests >>> showed it as quite fast. >>> >>> Any suggestions on why this could be happening >>> greatly appreciated. >>> >>> tested via >>> dd if=/dev/zero of=./testfile bs=16 count=16384 >> Gjournal is ~two times faster than UFS+SU for many >> small, random and >> parallel writes (such as running few 'tar x' >> processes), but is two >> times slower than UFS+SU for one sequential write >> stream, as there is no >> much that can be optimized there. >> >>> With disks getting larger and larger, why is it >>> taking so long for a journaled filesystem to be >>> standard on BSD? >> We have ZFS now, we don't need journaled file >> system:) > > > Here is an example of my tests and what I am > experiencing. > > FreeBSD server.old 6.2-PRERELEASE FreeBSD > 6.2-PRERELEASE #1: Wed Nov 22 21:20:08 PST 2006 > > With Gjournal: > root@:/1-journal> time dd if=/dev/zero > of=/1-journal/testfile1 bs=16k > count=16384 > 16384+0 records in > 16384+0 records out > 268435456 bytes transferred in 10.257279 secs > (26170240 bytes/sec) > dd if=/dev/zero of=/cache1/testfile1 bs=16k > count=16384 0.02s user 0.80s system 7% cpu 10.393 > total > > Without Gjournal: > root@1:/2-UFS> time dd if=/dev/zero of=/2/testfile1 > bs=16k > count=16384 > 16384+0 records in > 16384+0 records out > 268435456 bytes transferred in 26.682630 secs > (10060307 bytes/sec) > dd if=/dev/zero of=/home/testfile1 bs=16k count=16384 > 0.01s user 0.61s system 2% cpu 28.448 total > > As you can see, with this test, using Gjournaling > tested faster. > > Now on a new build: > > FreeBSD newtest 6.2-STABLE FreeBSD 6.2-STABLE #1: Tue > Aug 14 14:31:53 PDT 2007 > nicole@newtest:/usr/obj/spare/src/sys/GENERAL amd64 > > Without Gjournal > root@newtest:/home/nicole> time dd if=/dev/zero > of=./testfile bs=16k count=16384 > 16384+0 records in > 16384+0 records out > 268435456 bytes transferred in 4.749401 secs (56519854 > bytes/sec) > dd if=/dev/zero of=./testfile bs=16k count=16384 > 0.01s user 0.45s system 9% cpu 4.751 total > > With Gjournaling: > root@newtest:/home/nicole> time dd if=/dev/zero > of=/tmp/testfile bs=16k > count=16384 > 16384+0 records in > 16384+0 records out > 268435456 bytes transferred in 7.505348 secs (35765891 > bytes/sec) > dd if=/dev/zero of=/tmp/testfile bs=16k count=16384 > 0.00s user 0.57s system 7% cpu 7.517 total > > So as shown here, my test on an older 6.2 (with same > patches) showed improved performance. But now on a new > 6.2 build system, shows decreased performance. Could it be that you are not writing quite enough data to fully test it? Maybe in the 'old' server case, you were stopping the write *just* before the journal switch happened, so you happened to see the best sequential write throughput. The other tests could happen to fall across a journal switch, which would increase the time for the test, probably dropping the throughput in half.. What do you think? Eric From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 04:59:39 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 60B3816A417 for ; Sat, 18 Aug 2007 04:59:39 +0000 (UTC) (envelope-from tom.hurst@clara.net) Received: from spork.qfe3.net (spork.qfe3.net [212.13.207.101]) by mx1.freebsd.org (Postfix) with ESMTP id 203F613C45E for ; Sat, 18 Aug 2007 04:59:39 +0000 (UTC) (envelope-from tom.hurst@clara.net) Received: from [81.104.144.87] (helo=voi.aagh.net) by spork.qfe3.net with esmtp (Exim 4.66 (FreeBSD)) (envelope-from ) id 1IMG3K-0001uq-Tz; Sat, 18 Aug 2007 05:36:46 +0100 Received: from freaky by voi.aagh.net with local (Exim 4.67 (FreeBSD)) (envelope-from ) id 1IMG3K-000Pbh-KM; Sat, 18 Aug 2007 05:36:46 +0100 Date: Sat, 18 Aug 2007 05:36:46 +0100 From: Thomas Hurst To: Ivan Voras Message-ID: <20070818043646.GA96661@voi.aagh.net> Mail-Followup-To: Ivan Voras , freebsd-fs@freebsd.org References: <84373.93412.qm@web34511.mail.mud.yahoo.com> <20070817154139.GA1089@garage.freebsd.pl> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="pWyiEgJYm5f9v55/" Content-Disposition: inline In-Reply-To: Organization: Not much. User-Agent: Mutt/1.5.16 (2007-06-09) Sender: Thomas Hurst Cc: freebsd-fs@freebsd.org Subject: Re: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 04:59:39 -0000 --pWyiEgJYm5f9v55/ Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable * Ivan Voras (ivoras@fer.hr) wrote: > Pawel Jakub Dawidek wrote: >=20 > > We have ZFS now, we don't need journaled file system:) >=20 > I hope development on other file systems on FreeBSD won't stop because > of ZFS :) Choice is always good Quite. Personally I'm quite looking forward to Bluffs: http://asiabsdcon.org/papers//P11-slides.pdf Perhaps not as much as ZFS, but a more conservative option is always good. --=20 Thomas 'Freaky' Hurst http://hur.st/ --pWyiEgJYm5f9v55/ Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (FreeBSD) iD8DBQFGxndeNBBHZ542MwQRArDeAJ95oeoqJ9C5EZf6pUQ9/EWJ2w2vWgCcCLz6 A+h4g6MgKs5AHzKk5euEJNk= =TGuz -----END PGP SIGNATURE----- --pWyiEgJYm5f9v55/-- From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 08:09:26 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6744A16A417 for ; Sat, 18 Aug 2007 08:09:26 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: from web34502.mail.mud.yahoo.com (web34502.mail.mud.yahoo.com [66.163.178.168]) by mx1.freebsd.org (Postfix) with SMTP id 165FC13C459 for ; Sat, 18 Aug 2007 08:09:26 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: (qmail 1670 invoked by uid 60001); 18 Aug 2007 08:09:25 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=fc/l8LQKYtMXney7lE7jphKefV3ZdGzCSN22/MdyBBcoTDqDb4V7a55LYkrxIjCyRySoqqcC83FQP/8xCCvF1PZIkCoEdKF7ZAWiseHKD1cnO+Dm7wV2znMp0ipd+XOV4Kiq+GlGW+P9MCmyIH3c+QMF9t4JcbFmRmz87qzS2rs=; X-YMail-OSG: YMQv.ZAVM1muUFYfIYWsX4i394YNS3uLbb4cKSjzxXt90Rysi4Fs15tEN5mO_IQjooZJSToFuxxh4NzRM9.32n4kPzejCckMoUKJ6geygnV_1aeVkQI- Received: from [67.112.21.27] by web34502.mail.mud.yahoo.com via HTTP; Sat, 18 Aug 2007 01:09:24 PDT Date: Sat, 18 Aug 2007 01:09:24 -0700 (PDT) From: "N. Harrington" To: Thomas Hurst , Ivan Voras In-Reply-To: <20070818043646.GA96661@voi.aagh.net> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <330150.99914.qm@web34502.mail.mud.yahoo.com> Cc: freebsd-fs@freebsd.org Subject: Re: Gjournal reporting 1/2 the speed of non journaled? What is the status of Gjournal? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 08:09:26 -0000 --- Thomas Hurst wrote: > * Ivan Voras (ivoras@fer.hr) wrote: > > > Pawel Jakub Dawidek wrote: > > > > > We have ZFS now, we don't need journaled file > system:) > > > > I hope development on other file systems on > FreeBSD won't stop because > > of ZFS :) Choice is always good > > Quite. Personally I'm quite looking forward to > Bluffs: > > http://asiabsdcon.org/papers//P11-slides.pdf > > Perhaps not as much as ZFS, but a more conservative > option is always > good. > > -- > Thomas 'Freaky' Hurst > http://hur.st/ This seems intersesting. Any idea when it may be released into the wild? It says he hoped for Q2 release and we are now in Quarter 4. Does he mean 2nd quarter of next year? Thanks Nicole The Large Print Giveth And The Small Print Taketh Away -- Anon From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 08:20:45 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 1805216A419 for ; Sat, 18 Aug 2007 08:20:45 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: from web34505.mail.mud.yahoo.com (web34505.mail.mud.yahoo.com [66.163.178.171]) by mx1.freebsd.org (Postfix) with SMTP id DAA0C13C4CC for ; Sat, 18 Aug 2007 08:20:44 +0000 (UTC) (envelope-from drumslayer2@yahoo.com) Received: (qmail 10584 invoked by uid 60001); 18 Aug 2007 08:20:44 -0000 DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=s1024; d=yahoo.com; h=X-YMail-OSG:Received:Date:From:Subject:To:Cc:In-Reply-To:MIME-Version:Content-Type:Content-Transfer-Encoding:Message-ID; b=i7jKEzwAEhOtEQwlBR5qDM1E+IswJypwHXcmdU/Edx10SjsamVN6f4a6pVGoLO/6/oZQcCwjp6nb8o6Ezrvl3uE/i+KzhhjtT8eoipGRx1Eze4JBAMny93WOhhYOp5BQum3Q9W5RtGlhijwVbcFxtV8v4DUEjTvn/PhZ24oDR9c=; X-YMail-OSG: ooESDasVM1kj.njd3Vz4EWeeq29BcWcO80tqtc1jReN.smIH Received: from [67.112.21.27] by web34505.mail.mud.yahoo.com via HTTP; Sat, 18 Aug 2007 01:20:44 PDT Date: Sat, 18 Aug 2007 01:20:44 -0700 (PDT) From: "N. Harrington" To: Eric Anderson In-Reply-To: <46C675E3.1000001@freebsd.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Message-ID: <249926.8148.qm@web34505.mail.mud.yahoo.com> Cc: freebsd-fs@freebsd.org, Pawel Jakub Dawidek Subject: Re: Gjournal reporting 1/2 the speed of non journaled? - examples X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 08:20:45 -0000 --- Eric Anderson wrote: > N. Harrington wrote: > > --- Pawel Jakub Dawidek wrote: > > > >> On Thu, Aug 16, 2007 at 06:02:09PM -0700, N. > >> Harrington wrote: > >>> With ZFS, I have not seen much new going on > with > >>> gjournal. I am curious what the status of > gjournal > >> and > >>> if it will likely be included with 6.3 (whenever > >> that > >>> is due) > >> That was the plan, but I'm not yet sure if I'll > find > >> time to do it. > >> > >>> Also, as of late, I have been using it with > >>> 6.2-STABLE via the patches and I seem to be > >> getting > >>> 1/2 the transfer speeds compared to non > journaled > >>> disks. It seems like this is recent as previous > >> tests > >>> showed it as quite fast. > >>> > >>> Any suggestions on why this could be happening > >>> greatly appreciated. > >>> > >>> tested via > >>> dd if=/dev/zero of=./testfile bs=16 count=16384 > >> Gjournal is ~two times faster than UFS+SU for > many > >> small, random and > >> parallel writes (such as running few 'tar x' > >> processes), but is two > >> times slower than UFS+SU for one sequential write > >> stream, as there is no > >> much that can be optimized there. > >> > >>> With disks getting larger and larger, why is it > >>> taking so long for a journaled filesystem to be > >>> standard on BSD? > >> We have ZFS now, we don't need journaled file > >> system:) > > > > > > Here is an example of my tests and what I am > > experiencing. > > > > FreeBSD server.old 6.2-PRERELEASE FreeBSD > > 6.2-PRERELEASE #1: Wed Nov 22 21:20:08 PST 2006 > > > > With Gjournal: > > root@:/1-journal> time dd if=/dev/zero > > of=/1-journal/testfile1 bs=16k > > count=16384 > > 16384+0 records in > > 16384+0 records out > > 268435456 bytes transferred in 10.257279 secs > > (26170240 bytes/sec) > > dd if=/dev/zero of=/cache1/testfile1 bs=16k > > count=16384 0.02s user 0.80s system 7% cpu 10.393 > > total > > > > Without Gjournal: > > root@1:/2-UFS> time dd if=/dev/zero > of=/2/testfile1 > > bs=16k > > count=16384 > > 16384+0 records in > > 16384+0 records out > > 268435456 bytes transferred in 26.682630 secs > > (10060307 bytes/sec) > > dd if=/dev/zero of=/home/testfile1 bs=16k > count=16384 > > 0.01s user 0.61s system 2% cpu 28.448 total > > > > As you can see, with this test, using Gjournaling > > tested faster. > > > > Now on a new build: > > > > FreeBSD newtest 6.2-STABLE FreeBSD 6.2-STABLE #1: > Tue > > Aug 14 14:31:53 PDT 2007 > > nicole@newtest:/usr/obj/spare/src/sys/GENERAL > amd64 > > > > Without Gjournal > > root@newtest:/home/nicole> time dd if=/dev/zero > > of=./testfile bs=16k count=16384 > > 16384+0 records in > > 16384+0 records out > > 268435456 bytes transferred in 4.749401 secs > (56519854 > > bytes/sec) > > dd if=/dev/zero of=./testfile bs=16k count=16384 > > 0.01s user 0.45s system 9% cpu 4.751 total > > > > With Gjournaling: > > root@newtest:/home/nicole> time dd if=/dev/zero > > of=/tmp/testfile bs=16k > > count=16384 > > 16384+0 records in > > 16384+0 records out > > 268435456 bytes transferred in 7.505348 secs > (35765891 > > bytes/sec) > > dd if=/dev/zero of=/tmp/testfile bs=16k > count=16384 > > 0.00s user 0.57s system 7% cpu 7.517 total > > > > So as shown here, my test on an older 6.2 (with > same > > patches) showed improved performance. But now on a > new > > 6.2 build system, shows decreased performance. > > Could it be that you are not writing quite enough > data to fully test it? > Maybe in the 'old' server case, you were stopping > the write *just* > before the journal switch happened, so you happened > to see the best > sequential write throughput. The other tests could > happen to fall > across a journal switch, which would increase the > time for the test, > probably dropping the throughput in half.. What do > you think? > > > Eric > Sadly I do not think my testing is in error. Since I am comparing them equally and on very different systems. They are also very repeatable. Thus I wonder if some changes have occurred in 6.X within the past months that have caused gjournal not not integrate well. gjournal has not changed, since the patches have been the same for some time. Also another post I had on questions was how, for me, with 7.0 on a scsi system, it was reporting the disk speed as 160Mb/s instead of 320Mb/s as it would on 6.X. Speed tests show the same, but on bootup, the speeds are not reported the same. Perhaps the result of some of the changes being made on 7? If I get a chance, and figure out how, I will try to scan the change log for 6.X for any changes in the disk systems. I know of some changes that broke Pawel's patches in 2 places awhile ago. But I know even after those changes the speed was not effected. The speed decline seems recent. Be well. Nicole From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 10:27:09 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 1F47416A419 for ; Sat, 18 Aug 2007 10:27:09 +0000 (UTC) (envelope-from kvs@binarysolutions.dk) Received: from solow.pil.dk (relay.pil.dk [195.41.47.164]) by mx1.freebsd.org (Postfix) with ESMTP id E0BB813C459 for ; Sat, 18 Aug 2007 10:27:08 +0000 (UTC) (envelope-from kvs@binarysolutions.dk) Received: from coruscant.local (naboo.binarysolutions.dk [80.196.17.173]) by solow.pil.dk (Postfix) with ESMTP id D46E91CC0E8 for ; Sat, 18 Aug 2007 12:05:30 +0200 (CEST) Received: by coruscant.local (Postfix, from userid 502) id 0DCF6584A77; Sat, 18 Aug 2007 12:05:27 +0200 (CEST) To: freebsd-fs@freebsd.org From: Kenneth Vestergaard Schmidt Date: Sat, 18 Aug 2007 12:05:27 +0200 Message-ID: User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.1 (darwin) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Subject: ZFS: 'checksum mismatch' all over the place X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 10:27:09 -0000 Hello. We've just put a 12x750 GB raidz2 pool into use, but we're seeing constant 'checksum mismatch' errors. The drives are brand new. 'zpool status' currently lists the following: NAME STATE READ WRITE CKSUM pil ONLINE 0 0 189.9 raidz2 ONLINE 0 0 189.9 da0 ONLINE 0 0 2.99K da1 ONLINE 0 0 606 da2 ONLINE 0 0 75 da3 ONLINE 0 0 1.94K da4 ONLINE 0 0 786 da5 ONLINE 0 0 88 da6 ONLINE 0 0 79 da7 ONLINE 0 0 99 da8 ONLINE 0 0 533 da9 ONLINE 0 0 1.38K da10 ONLINE 0 0 15 da11 ONLINE 0 0 628 da0-da11 are really logical drives on an EonStor SCSI drive-cage. The physical disks are SATA, but since our EonStor can't run in JBOD-mode, I've had to create a logical drive per physical drive, and map each onto a separate SCSI LUN. The drive-cage was previously used to expose a RAID-5 array, composed of the 12 disks. This worked just fine, connecting to the same machine and controller (i386 IBM xSeries X335, mpt(4) controller). The EonStor can report SMART-statistics on each SATA-drive, and everything looks peachy there. What puzzles me is, that the drives don't seem to be failing - they just develop checksum errors. If they had hard failures, ZFS should mark them broken. It's also spread across all disks, and I have a hard time believing we just got 12 bad drives, which don't register as bad to the EonStor. Has anybody seen something like this? Any pointers on how to debug it? -- Kenneth Schmidt pil.dk From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 10:49:00 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 313DC16A420 for ; Sat, 18 Aug 2007 10:49:00 +0000 (UTC) (envelope-from des@des.no) Received: from tim.des.no (tim.des.no [194.63.250.121]) by mx1.freebsd.org (Postfix) with ESMTP id DC5E413C465 for ; Sat, 18 Aug 2007 10:48:59 +0000 (UTC) (envelope-from des@des.no) Received: from tim.des.no (localhost [127.0.0.1]) by spam.des.no (Postfix) with ESMTP id 353D020B0; Sat, 18 Aug 2007 12:48:54 +0200 (CEST) X-Spam-Tests: AWL X-Spam-Learn: disabled X-Spam-Score: 0.0/3.0 X-Spam-Checker-Version: SpamAssassin 3.2.1 (2007-05-02) on tim.des.no Received: from ds4.des.no (des.no [80.203.243.180]) by smtp.des.no (Postfix) with ESMTP id 50DA8208A; Sat, 18 Aug 2007 12:48:53 +0200 (CEST) Received: by ds4.des.no (Postfix, from userid 1001) id 1F8AC84479; Sat, 18 Aug 2007 12:48:53 +0200 (CEST) From: =?utf-8?Q?Dag-Erling_Sm=C3=B8rgrav?= To: rick-freebsd@kiwi-computer.com References: <723681.52797.qm@web63012.mail.re1.yahoo.com> <20070817020201.GA41414@keira.kiwi-computer.com> <867inuibu4.fsf@ds4.des.no> <20070817225103.GA92437@keira.kiwi-computer.com> Date: Sat, 18 Aug 2007 12:48:53 +0200 In-Reply-To: <20070817225103.GA92437@keira.kiwi-computer.com> (Rick C. Petty's message of "Fri\, 17 Aug 2007 17\:51\:03 -0500") Message-ID: <86d4xli0dm.fsf@ds4.des.no> User-Agent: Gnus/5.110006 (No Gnus v0.6) Emacs/22.1 (berkeley-unix) MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Cc: freebsd-fs@freebsd.org, Gore Jarold Subject: Re: vfs.ufs.dirhash_maxmem ... how high can I go ? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 10:49:00 -0000 "Rick C. Petty" writes: > Dag-Erling Sm=C3=B8rgrav writes: > > "Rick C. Petty" writes: > > > Gore Jarold writes: > > > > I have 4 GB of physical and 4 GB of swap, running on normal 32-bit > > > > x86, and I have this set as well: > > > Which means (on an x86 system) that you have 3 GB of physical RAM. > > No, it means has 4 GB of physical RAM, of which 3.5 GB are addressable. > I've never seen FreeBSD address more than 3.0 GB of RAM without PAE.. it > always seems to reserve 1.0 GB for video & other mmap'd I/O. FreeBSD does not reserve anything "for video & other mmap'd I/O". 512 MB of the 4 GB address space are reserved for PCI, but that is a feature of the IA32 architecture, not of FreeBSD. DES --=20 Dag-Erling Sm=C3=B8rgrav - des@des.no From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 12:20:33 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 6F4F916A419; Sat, 18 Aug 2007 12:20:33 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from ns.trinitel.com (186.161.36.72.static.reverse.ltdomains.com [72.36.161.186]) by mx1.freebsd.org (Postfix) with ESMTP id 39A5313C459; Sat, 18 Aug 2007 12:20:33 +0000 (UTC) (envelope-from anderson@freebsd.org) Received: from proton.local (r74-193-81-203.pfvlcmta01.grtntx.tl.dh.suddenlink.net [74.193.81.203]) (authenticated bits=0) by ns.trinitel.com (8.14.1/8.14.1) with ESMTP id l7ICKTTE058505; Sat, 18 Aug 2007 07:20:32 -0500 (CDT) (envelope-from anderson@freebsd.org) Message-ID: <46C6E40A.2040300@freebsd.org> Date: Sat, 18 Aug 2007 07:20:26 -0500 From: Eric Anderson User-Agent: Thunderbird 2.0.0.6 (Macintosh/20070728) MIME-Version: 1.0 To: "N. Harrington" References: <249926.8148.qm@web34505.mail.mud.yahoo.com> In-Reply-To: <249926.8148.qm@web34505.mail.mud.yahoo.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Spam-Status: No, score=-1.7 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.8 X-Spam-Checker-Version: SpamAssassin 3.1.8 (2007-02-13) on ns.trinitel.com Cc: freebsd-fs@freebsd.org, Pawel Jakub Dawidek Subject: Re: Gjournal reporting 1/2 the speed of non journaled? - examples X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 12:20:33 -0000 N. Harrington wrote: > --- Eric Anderson wrote: > >> N. Harrington wrote: >>> --- Pawel Jakub Dawidek wrote: >>> >>>> On Thu, Aug 16, 2007 at 06:02:09PM -0700, N. >>>> Harrington wrote: >>>>> With ZFS, I have not seen much new going on >> with >>>>> gjournal. I am curious what the status of >> gjournal >>>> and >>>>> if it will likely be included with 6.3 (whenever >>>> that >>>>> is due) >>>> That was the plan, but I'm not yet sure if I'll >> find >>>> time to do it. >>>> >>>>> Also, as of late, I have been using it with >>>>> 6.2-STABLE via the patches and I seem to be >>>> getting >>>>> 1/2 the transfer speeds compared to non >> journaled >>>>> disks. It seems like this is recent as previous >>>> tests >>>>> showed it as quite fast. >>>>> >>>>> Any suggestions on why this could be happening >>>>> greatly appreciated. >>>>> >>>>> tested via >>>>> dd if=/dev/zero of=./testfile bs=16 count=16384 >>>> Gjournal is ~two times faster than UFS+SU for >> many >>>> small, random and >>>> parallel writes (such as running few 'tar x' >>>> processes), but is two >>>> times slower than UFS+SU for one sequential write >>>> stream, as there is no >>>> much that can be optimized there. >>>> >>>>> With disks getting larger and larger, why is it >>>>> taking so long for a journaled filesystem to be >>>>> standard on BSD? >>>> We have ZFS now, we don't need journaled file >>>> system:) >>> >>> Here is an example of my tests and what I am >>> experiencing. >>> >>> FreeBSD server.old 6.2-PRERELEASE FreeBSD >>> 6.2-PRERELEASE #1: Wed Nov 22 21:20:08 PST 2006 >>> >>> With Gjournal: >>> root@:/1-journal> time dd if=/dev/zero >>> of=/1-journal/testfile1 bs=16k >>> count=16384 >>> 16384+0 records in >>> 16384+0 records out >>> 268435456 bytes transferred in 10.257279 secs >>> (26170240 bytes/sec) >>> dd if=/dev/zero of=/cache1/testfile1 bs=16k >>> count=16384 0.02s user 0.80s system 7% cpu 10.393 >>> total >>> >>> Without Gjournal: >>> root@1:/2-UFS> time dd if=/dev/zero >> of=/2/testfile1 >>> bs=16k >>> count=16384 >>> 16384+0 records in >>> 16384+0 records out >>> 268435456 bytes transferred in 26.682630 secs >>> (10060307 bytes/sec) >>> dd if=/dev/zero of=/home/testfile1 bs=16k >> count=16384 >>> 0.01s user 0.61s system 2% cpu 28.448 total >>> >>> As you can see, with this test, using Gjournaling >>> tested faster. >>> >>> Now on a new build: >>> >>> FreeBSD newtest 6.2-STABLE FreeBSD 6.2-STABLE #1: >> Tue >>> Aug 14 14:31:53 PDT 2007 >>> nicole@newtest:/usr/obj/spare/src/sys/GENERAL >> amd64 >>> Without Gjournal >>> root@newtest:/home/nicole> time dd if=/dev/zero >>> of=./testfile bs=16k count=16384 >>> 16384+0 records in >>> 16384+0 records out >>> 268435456 bytes transferred in 4.749401 secs >> (56519854 >>> bytes/sec) >>> dd if=/dev/zero of=./testfile bs=16k count=16384 >>> 0.01s user 0.45s system 9% cpu 4.751 total >>> >>> With Gjournaling: >>> root@newtest:/home/nicole> time dd if=/dev/zero >>> of=/tmp/testfile bs=16k >>> count=16384 >>> 16384+0 records in >>> 16384+0 records out >>> 268435456 bytes transferred in 7.505348 secs >> (35765891 >>> bytes/sec) >>> dd if=/dev/zero of=/tmp/testfile bs=16k >> count=16384 >>> 0.00s user 0.57s system 7% cpu 7.517 total >>> >>> So as shown here, my test on an older 6.2 (with >> same >>> patches) showed improved performance. But now on a >> new >>> 6.2 build system, shows decreased performance. >> Could it be that you are not writing quite enough >> data to fully test it? >> Maybe in the 'old' server case, you were stopping >> the write *just* >> before the journal switch happened, so you happened >> to see the best >> sequential write throughput. The other tests could >> happen to fall >> across a journal switch, which would increase the >> time for the test, >> probably dropping the throughput in half.. What do >> you think? >> >> >> Eric >> > Sadly I do not think my testing is in error. Since I > am comparing them equally and on very different > systems. They are also very repeatable. Thus I wonder > if some changes have occurred in 6.X within the past > months that have caused gjournal not not integrate > well. gjournal has not changed, since the patches have > been the same for some time. > > Also another post I had on questions was how, for me, > with 7.0 on a scsi system, it was reporting the disk > speed as 160Mb/s instead of 320Mb/s as it would on > 6.X. Speed tests show the same, but on bootup, the > speeds are not reported the same. Perhaps the result > of some of the changes being made on 7? > > If I get a chance, and figure out how, I will try to > scan the change log for 6.X for any changes in the > disk systems. I know of some changes that broke > Pawel's patches in 2 places awhile ago. But I know > even after those changes the speed was not effected. > The speed decline seems recent. Hmm.. Well, you could try the 'binary search' method, by cvs upping to a specific date, try it, and see if it exists, until you know that it happened in a certain range of time. Then finding the offending commit is *much* easier. If you narrow it down to a day or so, and post that, someone on this list can help you for sure. It's a time consuming task though... Eric From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 12:26:43 2007 Return-Path: Delivered-To: fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id CBD1916A418 for ; Sat, 18 Aug 2007 12:26:43 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from cyrus.watson.org (cyrus.watson.org [209.31.154.42]) by mx1.freebsd.org (Postfix) with ESMTP id A648B13C457 for ; Sat, 18 Aug 2007 12:26:43 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from fledge.watson.org (fledge.watson.org [209.31.154.41]) by cyrus.watson.org (Postfix) with ESMTP id 3AAE649441; Sat, 18 Aug 2007 07:58:14 -0400 (EDT) Date: Sat, 18 Aug 2007 12:58:14 +0100 (BST) From: Robert Watson X-X-Sender: robert@fledge.watson.org To: Jeff Mohler In-Reply-To: Message-ID: <20070818125647.O84677@fledge.watson.org> References: <46B0F505.8090102@freebsdbrasil.com.br> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: Patrick Tracanelli , fs@freebsd.org Subject: Re: Xsan (Apple) on FreeBSD X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 12:26:43 -0000 On Wed, 1 Aug 2007, Jeff Mohler wrote: > Im yet to hear of a large Xsan install that stayed Xsan once it grew. > > Most, if not all, have gone to netapp or umm..Isilon (spelling) that ive > been close to. Latest large dump of Xsan that I know of was Current TV in > San Francisco, for Isilon. > > Beware the hype of Xsan, its a money pit. > > Then ask em how long it all comes to a halt when a disk fails. Catching up on an aging thread here -- as far as I know, the XSan parts from Mac OS X are closed source, so while you can access XSan storage using whatever distributed file systems Apple supports (NFS, CIFS?), you can't use FreeBSD to directly access the storage area network. This is probably fine. You'll be interested to know, if you don't already, that both NetApp and Isilon use FreeBSD as the foundation OS for their products. Robert N M Watson Computer Laboratory University of Cambridge From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 13:55:43 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2CE6E16A417 for ; Sat, 18 Aug 2007 13:55:43 +0000 (UTC) (envelope-from pjd@garage.freebsd.pl) Received: from mail.garage.freebsd.pl (arm132.internetdsl.tpnet.pl [83.17.198.132]) by mx1.freebsd.org (Postfix) with ESMTP id C5A4E13C48D for ; Sat, 18 Aug 2007 13:55:42 +0000 (UTC) (envelope-from pjd@garage.freebsd.pl) Received: by mail.garage.freebsd.pl (Postfix, from userid 65534) id E52CF45CD9; Sat, 18 Aug 2007 15:55:40 +0200 (CEST) Received: from localhost (154.81.datacomsa.pl [195.34.81.154]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mail.garage.freebsd.pl (Postfix) with ESMTP id 946CF45684; Sat, 18 Aug 2007 15:55:33 +0200 (CEST) Date: Sat, 18 Aug 2007 15:54:38 +0200 From: Pawel Jakub Dawidek To: Ivan Voras Message-ID: <20070818135438.GC6498@garage.freebsd.pl> References: <20070817155741.GA6255@cons.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="Md/poaVZ8hnGTzuv" Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.4.2.3i X-PGP-Key-URL: http://people.freebsd.org/~pjd/pjd.asc X-OS: FreeBSD 7.0-CURRENT i386 X-Spam-Checker-Version: SpamAssassin 3.0.4 (2005-06-05) on mail.garage.freebsd.pl X-Spam-Level: X-Spam-Status: No, score=-2.6 required=3.0 tests=BAYES_00 autolearn=ham version=3.0.4 Cc: freebsd-fs@freebsd.org Subject: Re: Block device over network from Linux to FreeBSD X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 13:55:43 -0000 --Md/poaVZ8hnGTzuv Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Aug 17, 2007 at 08:50:26PM +0200, Ivan Voras wrote: > Martin Cracauer wrote: > > All right, here's a question that'll make your IQ drop by 5 points > > just from pondering it :-) >=20 > > Any other ideas? >=20 > If you're into programming, you can port the ggatec daemon that serves > as an end-point for ggated. Ggatec is userland-only and it just does IO > to/from a local file or device, nothing fancy (something like completely > nonstandard variant of iSCSI, ATA over ethernet, etc.). Right, but you confused ggatec with ggated. ggated is userland only daemon and is the one which should be ported. --=20 Pawel Jakub Dawidek http://www.wheel.pl pjd@FreeBSD.org http://www.FreeBSD.org FreeBSD committer Am I Evil? Yes, I Am! --Md/poaVZ8hnGTzuv Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQFGxvoeForvXbEpPzQRAj0kAKDA6Ot+pWVT4NA9gSSSSqIjgkmk9QCg8X6S V+1Lcaqwu8k568uIXfos6fI= =ItXx -----END PGP SIGNATURE----- --Md/poaVZ8hnGTzuv-- From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 14:26:10 2007 Return-Path: Delivered-To: fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id CE68116A418; Sat, 18 Aug 2007 14:26:10 +0000 (UTC) (envelope-from dlewis@solnetworks.net) Received: from que03.charter.net (que03.charter.net [209.225.8.191]) by mx1.freebsd.org (Postfix) with ESMTP id 21BD813C467; Sat, 18 Aug 2007 14:26:10 +0000 (UTC) (envelope-from dlewis@solnetworks.net) Received: from aa02.charter.net ([10.20.200.154]) by mtai05.charter.net (InterMail vM.7.08.02.00 201-2186-121-20061213) with ESMTP id <20070818140041.MOFQ7894.mtai05.charter.net@aa02.charter.net>; Sat, 18 Aug 2007 10:00:41 -0400 Received: from galileo ([24.176.104.6]) by aa02.charter.net with ESMTP id <20070818140041.SYZQ26124.aa02.charter.net@galileo>; Sat, 18 Aug 2007 10:00:41 -0400 Date: Sat, 18 Aug 2007 09:00:36 -0500 (CDT) From: "Derek E. Lewis" X-X-Sender: dlewis@galileo To: Robert Watson In-Reply-To: <20070818125647.O84677@fledge.watson.org> Message-ID: References: <46B0F505.8090102@freebsdbrasil.com.br> <20070818125647.O84677@fledge.watson.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed X-Chzlrs: 0 Cc: Patrick Tracanelli , fs@freebsd.org Subject: Re: Xsan (Apple) on FreeBSD X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 14:26:10 -0000 On Sat, 18 Aug 2007, Robert Watson wrote: > Catching up on an aging thread here -- as far as I know, the XSan parts from > Mac OS X are closed source, so while you can access XSan storage using > whatever distributed file systems Apple supports (NFS, CIFS?), you can't use > FreeBSD to directly access the storage area network. This is probably fine. > > You'll be interested to know, if you don't already, that both NetApp and > Isilon use FreeBSD as the foundation OS for their products. Isilon uses FreeBSD, yes, but Netapp uses Linux. One of the improvements Netapp made to Linux was rewriting the NFS stack to support NFSv4 in a decent manner. Those of you that have worked with Linux NFS before know that its not something you want to ship on a commercial storage product. Derek E. Lewis delewis@acm.org http://delewis.blogspot.com From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 14:38:58 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 49FEF16A418 for ; Sat, 18 Aug 2007 14:38:58 +0000 (UTC) (envelope-from freebsd-fs@m.gmane.org) Received: from ciao.gmane.org (main.gmane.org [80.91.229.2]) by mx1.freebsd.org (Postfix) with ESMTP id C6CDC13C480 for ; Sat, 18 Aug 2007 14:38:57 +0000 (UTC) (envelope-from freebsd-fs@m.gmane.org) Received: from list by ciao.gmane.org with local (Exim 4.43) id 1IMPS0-0007el-BN for freebsd-fs@freebsd.org; Sat, 18 Aug 2007 16:38:52 +0200 Received: from 78-1-112-132.adsl.net.t-com.hr ([78.1.112.132]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 18 Aug 2007 16:38:52 +0200 Received: from ivoras by 78-1-112-132.adsl.net.t-com.hr with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Sat, 18 Aug 2007 16:38:52 +0200 X-Injected-Via-Gmane: http://gmane.org/ To: freebsd-fs@freebsd.org From: Ivan Voras Date: Sat, 18 Aug 2007 16:38:35 +0200 Lines: 44 Message-ID: References: <20070817155741.GA6255@cons.org> <20070818135438.GC6498@garage.freebsd.pl> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enig9AAFEF284389218C657A8ECB" X-Complaints-To: usenet@sea.gmane.org X-Gmane-NNTP-Posting-Host: 78-1-112-132.adsl.net.t-com.hr User-Agent: Thunderbird 2.0.0.6 (Windows/20070728) In-Reply-To: <20070818135438.GC6498@garage.freebsd.pl> X-Enigmail-Version: 0.95.3 Sender: news Subject: Re: Block device over network from Linux to FreeBSD X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 14:38:58 -0000 This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enig9AAFEF284389218C657A8ECB Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Pawel Jakub Dawidek wrote: > On Fri, Aug 17, 2007 at 08:50:26PM +0200, Ivan Voras wrote: >> Martin Cracauer wrote: >>> All right, here's a question that'll make your IQ drop by 5 points >>> just from pondering it :-) >>> Any other ideas? >> If you're into programming, you can port the ggatec daemon that serves= >> as an end-point for ggated. Ggatec is userland-only and it just does I= O >> to/from a local file or device, nothing fancy (something like complete= ly >> nonstandard variant of iSCSI, ATA over ethernet, etc.). >=20 > Right, but you confused ggatec with ggated. ggated is userland only > daemon and is the one which should be ported. Yes, you're right, thanks! Everything above is correct, just reverse ggatec and ggated (c=3Dclient/consumer=3Dthe part that manages the kernel device, d=3Ddaemon/provider=3Dthe part that provides the data for the device). --------------enig9AAFEF284389218C657A8ECB Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.5 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGxwRxldnAQVacBcgRAphZAJ9lQP+qgYOXkRjSU6uczEuXz2mV+wCfdrGR nuYWHyQVTUhnVGqF9UZW2sw= =eeRj -----END PGP SIGNATURE----- --------------enig9AAFEF284389218C657A8ECB-- From owner-freebsd-fs@FreeBSD.ORG Sat Aug 18 14:46:18 2007 Return-Path: Delivered-To: fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id CCF1716A418 for ; Sat, 18 Aug 2007 14:46:18 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from cyrus.watson.org (cyrus.watson.org [209.31.154.42]) by mx1.freebsd.org (Postfix) with ESMTP id 7E82813C458 for ; Sat, 18 Aug 2007 14:46:18 +0000 (UTC) (envelope-from rwatson@FreeBSD.org) Received: from fledge.watson.org (fledge.watson.org [209.31.154.41]) by cyrus.watson.org (Postfix) with ESMTP id 64373494B3; Sat, 18 Aug 2007 10:46:17 -0400 (EDT) Date: Sat, 18 Aug 2007 15:46:17 +0100 (BST) From: Robert Watson X-X-Sender: robert@fledge.watson.org To: "Derek E. Lewis" In-Reply-To: Message-ID: <20070818154250.V27632@fledge.watson.org> References: <46B0F505.8090102@freebsdbrasil.com.br> <20070818125647.O84677@fledge.watson.org> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Cc: Patrick Tracanelli , fs@freebsd.org Subject: Re: Xsan (Apple) on FreeBSD X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 18 Aug 2007 14:46:18 -0000 On Sat, 18 Aug 2007, Derek E. Lewis wrote: > On Sat, 18 Aug 2007, Robert Watson wrote: > >> Catching up on an aging thread here -- as far as I know, the XSan parts >> from Mac OS X are closed source, so while you can access XSan storage using >> whatever distributed file systems Apple supports (NFS, CIFS?), you can't >> use FreeBSD to directly access the storage area network. This is probably >> fine. >> >> You'll be interested to know, if you don't already, that both NetApp and >> Isilon use FreeBSD as the foundation OS for their products. > > Isilon uses FreeBSD, yes, but Netapp uses Linux. One of the improvements > Netapp made to Linux was rewriting the NFS stack to support NFSv4 in a > decent manner. Those of you that have worked with Linux NFS before know that > its not something you want to ship on a commercial storage product. NetApp gave a rather nice presentation at the recent FreeBSD developer summit in Ottawa on the topic of FreeBSD as the foundation OS for OnTap/GX, and also made a rather healthy donation to the FreeBSD Foundation in the last six months. I defer to their expertise on the point of what the OS in their product is... :-) As I understand it, NetApp has improved the Linux NFS client significantly, but not for the purposes of including it in their product. Robert N M Watson Computer Laboratory University of Cambridge