From owner-freebsd-fs Mon Feb 9 01:57:41 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id BAA29457 for freebsd-fs-outgoing; Mon, 9 Feb 1998 01:57:41 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id BAA29415; Mon, 9 Feb 1998 01:57:28 -0800 (PST) (envelope-from michaelh@cet.co.jp) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.8/CET-v2.2) with SMTP id JAA02398; Mon, 9 Feb 1998 09:56:39 GMT Date: Mon, 9 Feb 1998 18:56:38 +0900 (JST) From: Michael Hancock To: "Alton, Matthew" cc: fs@FreeBSD.ORG, hackers@FreeBSD.ORG Subject: Re: Filesystem hacking In-Reply-To: <31B3F0BF1C40D11192A700805FD48BF9C33287@STLABCEXG011> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 30 Jan 1998, Alton, Matthew wrote: > I have decided to code up an IBM-style journaling filesystem > (jfs) with maximum portability for free unices. While I'm at it Cool. Portability might be a challenge though, the vnode interface varies quite a bit between the various Unices. > I had might as well have the clean-bits map to a PP/extent > disk arrangement which will act as a useful abstraction for a > Logical Volume Manager / Veritas -esque disk management > system which I also find interesting enough to code up. > > This is not an attempt to morph FreeBSD into AIX by any > means. It is just an interesting project which I think may be > of use to hackers. > > This month (02/98) I will explore the GNU HURD OS to see if > their goal of creating an OS which would allow me to imple- > ment a filesystem in user space has been realized. If it has > not, I will use my normal FreeBSD 2.2.2 i386/40 and NetBSD > 1.3 Sun 3/50 & 3/80 development boxes. See the link below for a description of stacking, I was one of the people that was talking about doing ktou/utok layers in FBSD but I haven't gotten around to it yet. http://www.isi.edu/~johnh/SOFTWARE/UCLA_STACKING/ Doing a terminal layer in userland would be a pretty good challenge. While GNU HURD is interesting, for a production environment you should probably stick with doing most of it in the kernel. You might also want to look at the CryptoFS, I think it's in the ports tree, for an example of an FS that piggybacks on NFS. > Please let me know if I am duplicating effort and provide me > with relevant pointers. > P.S. it's good to see a FreeBSDer from AB. Any chance of converting Scott Smallie? ;-) -- michaelh@cet.co.jp http://www.cet.co.jp CET Inc., Daiichi Kasuya BLDG 8F, 2-5-12 Higashi Shinbashi, Minato-ku, Tokyo 105 Japan Tel: +81-3-3437-1761 Fax: +81-3-3437-1766 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Mon Feb 9 16:42:43 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA04280 for freebsd-fs-outgoing; Mon, 9 Feb 1998 16:42:43 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from iconetworks.com (psycho2.iconetworks.net [204.94.129.66]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA04275 for ; Mon, 9 Feb 1998 16:42:42 -0800 (PST) (envelope-from roark_hilomen_at_meridian@meridian-data.com) From: roark_hilomen_at_meridian@meridian-data.com Received: from meridian-data.com (cdserv.meridian-data.com [204.94.131.20]) by ICONETWORKS (8.8.5/8.8.2/103196) with SMTP id QAA12114 for ; Mon, 9 Feb 1998 16:42:40 -0800 Received: from ccMail by meridian-data.com (ccMail Link to SMTP R8.00.01) id AA887071056; Mon, 09 Feb 98 16:38:08 -0800 Message-Id: <9802098870.AA887071056@meridian-data.com> X-Mailer: ccMail Link to SMTP R8.00.01 Date: Mon, 09 Feb 98 16:43:15 -0800 To: Subject: Soft update usage MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Does anyone know if "Soft Updates" has been in use in the industry or is OpenBSD and now FreeBSD the first to implement it? How extensively has it been tested (outside the current FreeBSD flurry)? And for that matter, is anyone testing it on a large multi-client system? Has anyone tried crashing it and seeing if FSCK works well and as advertised with Soft Updates? I'm trying to see how cutting edge this technology is. Thanks, Roark To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 00:29:44 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA09039 for freebsd-fs-outgoing; Tue, 10 Feb 1998 00:29:44 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from smtp02.primenet.com (smtp02.primenet.com [206.165.6.132]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id AAA09030 for ; Tue, 10 Feb 1998 00:29:42 -0800 (PST) (envelope-from tlambert@usr05.primenet.com) Received: (from daemon@localhost) by smtp02.primenet.com (8.8.8/8.8.8) id BAA24768; Tue, 10 Feb 1998 01:29:38 -0700 (MST) Received: from usr05.primenet.com(206.165.6.205) via SMTP by smtp02.primenet.com, id smtpd024761; Tue Feb 10 01:29:32 1998 Received: (from tlambert@localhost) by usr05.primenet.com (8.8.5/8.8.5) id BAA21865; Tue, 10 Feb 1998 01:29:23 -0700 (MST) From: Terry Lambert Message-Id: <199802100829.BAA21865@usr05.primenet.com> Subject: Re: Soft update usage To: roark_hilomen_at_meridian@meridian-data.com Date: Tue, 10 Feb 1998 08:29:23 +0000 (GMT) Cc: freebsd-fs@FreeBSD.ORG In-Reply-To: <9802098870.AA887071056@meridian-data.com> from "roark_hilomen_at_meridian@meridian-data.com" at Feb 9, 98 04:43:15 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Does anyone know if "Soft Updates" has been in use in the industry > or is OpenBSD and now FreeBSD the first to implement it? BSDI has them. The original Ganger/Patt work was done in SVR4 (if you are familiar with SVR4 internals, this is obvious from looking at the published Appendix A code). > How extensively has it been tested (outside the current FreeBSD flurry)? Kirk would be a better person to answer that. > And for that matter, is anyone testing it on a large multi-client system? I don't know; they should. > Has anyone tried crashing it and seeing if FSCK works well and as advertised > with Soft Updates? That's Kirk's testbed. He uses a SCSI drive and two machines; he halts the one from the other, randomly, and fsck's (among other tests). He claims many, many months of reliable operation. > I'm trying to see how cutting edge this technology is. It's "medium". It's a good leap past current implementations, but the paper was published several years ago, and it does not support general updates between layered FS's. Effectively, Ganger and Patt did this: 1) Determine all operations that could be done on SVR4 UFS 2) Determine the events that would result from these operations, as small DAG fragments (directed acylic graphs). 3) Invent a structure called a "dependency" that defined relationships between these DAG's. 4) Invent a mechanism for ensuring ordering of operations in dependency order (this mechanism is superior to DOW -- delayed ordered writes -- a USL patent-pending technology, because there is, in effect, write gathering and meta-data ordering without flush -- unlike DOW). 5) Hard-code (rather than parametrically code) these dependency queueing events. This is the step that degeneralizes the Ganger/Patt code, and makes it a lot of work to port -- at a 3-5% performance increase (negligible in comparison to the win of soft updates themselves over using sync writes to guarantee ordering of operations). The DOW mechanism is basically the classic "banker's algorithm", which determines potential cycles in a graph and denies operations based on the possibility of cycles. The soft updates mechanism (effectively) precomputes transitive closure over the graph, so any insertion that would result in a cycle is seen as depending on prior insertions. The difference in these mechisms can be best characterized by: DOW: Prevent operations which are dangerous. Also, using back-tracking via Howard's algortihm, prevent operations which *may* be dangerous, even if it ends up that they *aren't* dangerous. SU: Prevent only operations which are dangerous. I *seriously* recommend the following books: Algorithms in C++ Robert Sedgewick Addison-Wesley Publishing Company ISBN 0-201-51059-6 The Art Of computer Programming _Volume 1: Fundamental Algorithms_ Donald Knuth Addison-Wesley Publishing Company ISBN 0-201-03809-9 UNIX Systems for Modern Architectures _Symmetric Multiprocessing and Caching for Kernel Programmers_ Curt Schimmel Addison-Wesley Publishing Company ISBN 0-201-63338-8 For more information on graphical algorithms. Don't be put off by the Sedgewick book; it is very light on the code and heavy on the graph theory (chapters 29 through 34 -- I recommend chapters 24 through 28, on geomentric algorithms, as well). For example, if you were to implement a lock manager using graphical representations and cycle detection, you could probably expect to do about 20,000 transactions a second (this is the measured rate through the NetWare for UNIX lock manager, which used an intention mode lock manager based on graphical algorithms). Of course, you may not want to use this in an SMP kernel, since it's only two orders of magnitude faster than Tuxedo, which used the same algorithm as the BSD lockmgr code. If you used a graphical soloution to locking in an SMP kernel, you'd have to put up with your kernel being faster, more highly concurrent, and unfortunately scalable to a larger number of processors than SVR4 or Solaris could every hope to support. Then where would you be? Oh. Yeah. You'd be at the cutting edge. Couldn't have that... Heh. 8-). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 05:44:30 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id FAA28210 for freebsd-fs-outgoing; Tue, 10 Feb 1998 05:44:30 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from inet-user-gw-1.us.oracle.com (inet-user-gw-1.us.oracle.com [192.86.155.82]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id FAA28203 for ; Tue, 10 Feb 1998 05:44:24 -0800 (PST) (envelope-from HARAWAT.IN.oracle.com.ofcmail@in.oracle.com) Received: from dwarpal.in.oracle.com (dwarpal.in.oracle.com [152.69.176.11]) by inet-user-gw-1.us.oracle.com (8.8.5/8.8.5) with SMTP id FAA19464 for ; Tue, 10 Feb 1998 05:42:06 -0800 (PST) Received: by dwarpal.in.oracle.com (8.6.13/37.8) id JAA04546; Tue, 10 Feb 1998 09:16:34 -0500 Message-Id: <199802101416.JAA04546@dwarpal.in.oracle.com> Date: 10 Feb 98 16:56:39 +0330 From: "HARAWAT.IN.ORACLE.COM" To: freebsd-fs@FreeBSD.ORG Subject: RAID support MIME-Version: 1.0 X-Mailer: Oracle InterOffice (version 4.0.5.1.55) Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Is RAID support available in FreeBSD. If yes then what all levels of RAID does freebsd support . Harish ----------------------------------------------------------------------------- Harish Singh Rawat E-Mail : harawat@in.oracle.com Oracle Software India. Ltd. Voice : +91 (80) 228 3004/5 Indian Development Center ( IDC ) Fax : +91 (80) 225 - 6207/1591 Sona Towers No 71 Miller Road Bangalore-52 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 06:47:18 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id GAA06504 for freebsd-fs-outgoing; Tue, 10 Feb 1998 06:47:18 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from leviathan.ele.uri.edu (leviathan.ele.uri.edu [131.128.51.64]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id GAA06499 for ; Tue, 10 Feb 1998 06:47:16 -0800 (PST) (envelope-from hu@ele.uri.edu) Received: from hull.ele.uri.edu (hull [131.128.51.149]) by leviathan.ele.uri.edu (8.8.8/8.8.8) with ESMTP id JAA23451 for ; Tue, 10 Feb 1998 09:47:14 -0500 (EST) From: Yiming Hu Received: (from hu@localhost) by hull.ele.uri.edu (8.8.8/8.8.8) id JAA00821 for freebsd-fs@freebsd.org; Tue, 10 Feb 1998 09:47:14 -0500 (EST) Message-Id: <199802101447.JAA00821@hull.ele.uri.edu> Subject: Re: Soft update usage To: freebsd-fs@FreeBSD.ORG Date: Tue, 10 Feb 1998 09:47:14 -0500 (EST) MIME-Version: 1.0 Content-Type: text/plain; charset=Chinese-GB Content-Transfer-Encoding: 8 bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Hi, Can anyone tell me where can I find more information (such as published papers or web sites) about DOW (Delayed Ordered Writes)? Thanks! -Yiming Hu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 07:37:35 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA12943 for freebsd-fs-outgoing; Tue, 10 Feb 1998 07:37:35 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from Kitten.mcs.com (Kitten.mcs.com [192.160.127.90]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA12938 for ; Tue, 10 Feb 1998 07:37:32 -0800 (PST) (envelope-from nash@Jupiter.Mcs.Net) Received: from Jupiter.Mcs.Net (nash@Jupiter.mcs.net [192.160.127.88]) by Kitten.mcs.com (8.8.7/8.8.2) with ESMTP id JAA10574; Tue, 10 Feb 1998 09:37:31 -0600 (CST) Received: from localhost (nash@localhost) by Jupiter.Mcs.Net (8.8.7/8.8.2) with SMTP id JAA07252; Tue, 10 Feb 1998 09:37:30 -0600 (CST) Date: Tue, 10 Feb 1998 09:37:30 -0600 (CST) From: Alex Nash To: Yiming Hu cc: freebsd-fs@FreeBSD.ORG Subject: Re: Soft update usage In-Reply-To: <199802101447.JAA00821@hull.ele.uri.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 10 Feb 1998, Yiming Hu wrote: > Can anyone tell me where can I find more information (such as > published papers or web sites) about DOW (Delayed Ordered Writes)? http://www.ece.cmu.edu/~ganger/papers/ Alex To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 07:53:02 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA14791 for freebsd-fs-outgoing; Tue, 10 Feb 1998 07:53:02 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from sumatra.americantv.com (sumatra.americantv.com [207.170.17.37]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA14780 for ; Tue, 10 Feb 1998 07:52:56 -0800 (PST) (envelope-from jlemon@americantv.com) Received: from right.PCS (right.PCS [148.105.10.31]) by sumatra.americantv.com (8.8.5/8.8.5) with ESMTP id JAA08020; Tue, 10 Feb 1998 09:52:48 -0600 (CST) Received: (from jlemon@localhost) by right.PCS (8.6.13/8.6.4) id JAA27979; Tue, 10 Feb 1998 09:52:17 -0600 Message-ID: <19980210095216.25074@right.PCS> Date: Tue, 10 Feb 1998 09:52:16 -0600 From: Jonathan Lemon To: Yiming Hu Cc: freebsd-fs@FreeBSD.ORG Subject: Re: Soft update usage References: <199802101447.JAA00821@hull.ele.uri.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.61.1 In-Reply-To: <199802101447.JAA00821@hull.ele.uri.edu>; from Yiming Hu on Feb 02, 1998 at 09:47:14AM -0500 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Feb 02, 1998 at 09:47:14AM -0500, Yiming Hu wrote: > Can anyone tell me where can I find more information (such as > published papers or web sites) about DOW (Delayed Ordered Writes)? G.R.Ganger, Y.N.Patt. Metadata Update Performance in File Systems. Operating Systems Design and Implementation, pp 49-60, November 1994. http://www.ece.cmu.edu/~ganger/papers/osdi94.ps G.R.Ganger, Y.N.Patt. Soft Updates: A Solution to the Metadata Update Problem in File Systems. University of Michigan Technical Report CSE-TR-254-95, August 1995. http://www.ece.cmu.edu/~ganger/papers/CSE-TR-254-95/ -- Jonathan To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 12:03:42 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA04162 for freebsd-fs-outgoing; Tue, 10 Feb 1998 12:03:42 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from smtp04.primenet.com (smtp04.primenet.com [206.165.6.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA04128 for ; Tue, 10 Feb 1998 12:03:40 -0800 (PST) (envelope-from tlambert@usr04.primenet.com) Received: (from daemon@localhost) by smtp04.primenet.com (8.8.8/8.8.8) id NAA14493; Tue, 10 Feb 1998 13:03:28 -0700 (MST) Received: from usr04.primenet.com(206.165.6.204) via SMTP by smtp04.primenet.com, id smtpd014470; Tue Feb 10 13:03:25 1998 Received: (from tlambert@localhost) by usr04.primenet.com (8.8.5/8.8.5) id NAA02397; Tue, 10 Feb 1998 13:03:24 -0700 (MST) From: Terry Lambert Message-Id: <199802102003.NAA02397@usr04.primenet.com> Subject: Re: RAID support To: HARAWAT.IN.oracle.com.ofcmail@in.oracle.com (HARAWAT.IN.ORACLE.COM) Date: Tue, 10 Feb 1998 20:03:24 +0000 (GMT) Cc: freebsd-fs@FreeBSD.ORG In-Reply-To: <199802101416.JAA04546@dwarpal.in.oracle.com> from "HARAWAT.IN.ORACLE.COM" at Feb 10, 98 04:56:39 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > Is RAID support available in FreeBSD. If yes then what all levels of > RAID does freebsd support . Yes. I posted patchees back to the maintainers (and the -current list) for the RAIDFrame code. There are also several RAID SCSI controllers that have drivers (which are not integrated into the main FreeBSD code base, including one for the Compaq controller). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 12:12:50 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA05926 for freebsd-fs-outgoing; Tue, 10 Feb 1998 12:12:50 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from pluto.plutotech.com (mail.plutotech.com [206.168.67.137]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA05913 for ; Tue, 10 Feb 1998 12:12:49 -0800 (PST) (envelope-from gibbs@plutotech.com) Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by pluto.plutotech.com (8.8.7/8.8.5) with ESMTP id NAA27515; Tue, 10 Feb 1998 13:12:41 -0700 (MST) Message-Id: <199802102012.NAA27515@pluto.plutotech.com> X-Mailer: exmh version 2.0.1 12/23/97 To: Terry Lambert cc: HARAWAT.IN.oracle.com.ofcmail@in.oracle.com (HARAWAT.IN.ORACLE.COM), freebsd-fs@FreeBSD.ORG Subject: Re: RAID support In-reply-to: Your message of "Tue, 10 Feb 1998 20:03:24 GMT." <199802102003.NAA02397@usr04.primenet.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 10 Feb 1998 13:09:56 -0700 From: "Justin T. Gibbs" Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org >Yes. I posted patchees back to the maintainers (and the -current list) >for the RAIDFrame code. There are also several RAID SCSI controllers >that have drivers (which are not integrated into the main FreeBSD >code base, including one for the Compaq controller). There are only two controllers that I am aware of that have driver support: the dpt, and the EISA Compaq controller. The dpt driver has been integrated. The Compaq driver has not due to some question about the legality of distributing it in source form. > Terry Lambert > terry@lambert.org -- Justin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 12:48:42 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA11070 for freebsd-fs-outgoing; Tue, 10 Feb 1998 12:48:42 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from galileo.orbinet.com.mx (galileo.orbinet.com.mx [200.33.130.242]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA11065 for ; Tue, 10 Feb 1998 12:48:40 -0800 (PST) (envelope-from erogelio@galileo.orbinet.com.mx) Received: from localhost (erogelio@localhost) by galileo.orbinet.com.mx (8.8.7/8.8.7) with SMTP id OAA10138; Tue, 10 Feb 1998 14:46:04 -0600 (CST) (envelope-from erogelio@galileo.orbinet.com.mx) Date: Tue, 10 Feb 1998 14:46:04 -0600 (CST) From: "Edgar Roberto Rogelio Mtz." To: Alex Nash cc: Yiming Hu , freebsd-fs@FreeBSD.ORG In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org i have a dude, about the unit tape, i don't know about the dispositive and i have to use it, for exameple i have a floppy disk and i mounted that something like this: mount -t msdos /dev/fd0 /mnt =) Somebody can help me??? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 13:11:24 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA15445 for freebsd-fs-outgoing; Tue, 10 Feb 1998 13:11:24 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from smtp02.primenet.com (smtp02.primenet.com [206.165.6.132]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA15434 for ; Tue, 10 Feb 1998 13:11:20 -0800 (PST) (envelope-from tlambert@usr06.primenet.com) Received: (from daemon@localhost) by smtp02.primenet.com (8.8.8/8.8.8) id OAA06188; Tue, 10 Feb 1998 14:11:09 -0700 (MST) Received: from usr06.primenet.com(206.165.6.206) via SMTP by smtp02.primenet.com, id smtpd006143; Tue Feb 10 14:11:00 1998 Received: (from tlambert@localhost) by usr06.primenet.com (8.8.5/8.8.5) id OAA04484; Tue, 10 Feb 1998 14:10:57 -0700 (MST) From: Terry Lambert Message-Id: <199802102110.OAA04484@usr06.primenet.com> Subject: Re: Soft update usage To: nash@Mcs.Net (Alex Nash) Date: Tue, 10 Feb 1998 21:10:57 +0000 (GMT) Cc: hu@ele.uri.edu, freebsd-fs@FreeBSD.ORG In-Reply-To: from "Alex Nash" at Feb 10, 98 09:37:30 am X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > > Can anyone tell me where can I find more information (such as > > published papers or web sites) about DOW (Delayed Ordered Writes)? > > http://www.ece.cmu.edu/~ganger/papers/ That's Soft Updates. You really don't want DOW, but I believe there is a research report out of the University of Kentucky "Choices" projects, which was supported by USL. All of the documents I saw were internal USL documents, so I can't disclose anything that wasn't obvious from the header files. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 13:43:38 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA20531 for freebsd-fs-outgoing; Tue, 10 Feb 1998 13:43:38 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from iconetworks.com (psycho2.iconetworks.net [204.94.129.66]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA20488 for ; Tue, 10 Feb 1998 13:43:32 -0800 (PST) (envelope-from roark_hilomen_at_meridian@meridian-data.com) From: roark_hilomen_at_meridian@meridian-data.com Received: from meridian-data.com (cdserv.meridian-data.com [204.94.131.20]) by ICONETWORKS (8.8.5/8.8.2/103196) with SMTP id NAA03923 for ; Tue, 10 Feb 1998 13:43:25 -0800 Received: from ccMail by meridian-data.com (ccMail Link to SMTP R8.00.01) id AA887143102; Tue, 10 Feb 98 13:38:56 -0800 Message-Id: <9802108871.AA887143102@meridian-data.com> X-Mailer: ccMail Link to SMTP R8.00.01 Date: Tue, 10 Feb 98 12:43:59 -0800 To: Subject: Any Performance Numbers MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Does anyone have performance numbers? How fast is Softupdate vs DOW vs Standard Clustered FFS (and vs LFS)? How fast does FSCK for Softupdate run vs FSCK for normal FFS (ganger paper mentions an 850Meg disk taking 5-7 minutes under normal FFS and 5+ seconds under Softupdate...has this been validated)? Is the user perceived increase significant? Thanks, Roark (Sorry for all the questions, but I am really trying to persuade myself to take the leap). To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 13:58:12 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id NAA23463 for freebsd-fs-outgoing; Tue, 10 Feb 1998 13:58:12 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id NAA23450 for ; Tue, 10 Feb 1998 13:58:08 -0800 (PST) (envelope-from grog@lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.8.7/8.8.5) with ESMTP id IAA16374; Wed, 11 Feb 1998 08:26:23 +1030 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.8.8/8.8.7) id IAA18934; Wed, 11 Feb 1998 08:26:23 +1030 (CST) (envelope-from grog) Message-ID: <19980211082623.02806@freebie.lemis.com> Date: Wed, 11 Feb 1998 08:26:23 +1030 From: Greg Lehey To: "HARAWAT.IN.ORACLE.COM" , freebsd-fs@FreeBSD.ORG Subject: Re: RAID support References: <199802101416.JAA04546@dwarpal.in.oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: <199802101416.JAA04546@dwarpal.in.oracle.com>; from HARAWAT.IN.ORACLE.COM on Tue, Feb 10, 1998 at 04:56:39PM +0330 WWW-Home-Page: http://www.lemis.com/~grog Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 10 February 1998 at 16:56:39 +0330, HARAWAT.IN.ORACLE.COM wrote: > Is RAID support available in FreeBSD. If yes then what all levels of > RAID does freebsd support . I'm currently working on a driver which will provide RAID 1 and RAID 5. Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 14:07:18 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id OAA25776 for freebsd-fs-outgoing; Tue, 10 Feb 1998 14:07:18 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id OAA25763 for ; Tue, 10 Feb 1998 14:07:14 -0800 (PST) (envelope-from grog@lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.8.7/8.8.5) with ESMTP id IAA16388; Wed, 11 Feb 1998 08:35:43 +1030 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.8.8/8.8.7) id IAA18985; Wed, 11 Feb 1998 08:35:21 +1030 (CST) (envelope-from grog) Message-ID: <19980211083521.07336@freebie.lemis.com> Date: Wed, 11 Feb 1998 08:35:21 +1030 From: Greg Lehey To: "Edgar Roberto Rogelio Mtz." , Alex Nash Cc: Yiming Hu , freebsd-fs@FreeBSD.ORG Subject: Re: your mail References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: ; from Edgar Roberto Rogelio Mtz. on Tue, Feb 10, 1998 at 02:46:04PM -0600 WWW-Home-Page: http://www.lemis.com/~grog Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Tue, 10 February 1998 at 14:46:04 -0600, Edgar Roberto Rogelio Mtz. wrote: > > > > i have a dude, about the unit tape, i don't know about the dispositive > and i have to use it, for exameple i have a floppy disk and i mounted that > something like this: > > mount -t msdos /dev/fd0 /mnt > > > =) > > Somebody can help me??? OK. How's this for a start: 1. If you have questions about using FreeBSD, address them to FreeBSD-questions@FreeBSD.org. 2. Don't follow up on some other topic when doing so. 3. Supply a subject line, or nobody will read your message. 4. Describe what you want to do. 5. Read http://www.lemis.com/questions.html for further suggestions. Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 16:50:46 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id QAA19888 for freebsd-fs-outgoing; Tue, 10 Feb 1998 16:50:46 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id QAA19883 for ; Tue, 10 Feb 1998 16:50:44 -0800 (PST) (envelope-from dyson@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id TAA07191; Tue, 10 Feb 1998 19:50:33 -0500 (EST) (envelope-from dyson) Message-Id: <199802110050.TAA07191@dyson.iquest.net> Subject: Re: Any Performance Numbers In-Reply-To: <9802108871.AA887143102@meridian-data.com> from "roark_hilomen_at_meridian@meridian-data.com" at "Feb 10, 98 12:43:59 pm" To: roark_hilomen_at_meridian@meridian-data.com Date: Tue, 10 Feb 1998 19:50:33 -0500 (EST) Cc: freebsd-fs@FreeBSD.ORG From: "John S. Dyson" Reply-To: dyson@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org roark_hilomen_at_meridian@meridian-data.com said: > > Does anyone have performance numbers? > > How fast is Softupdate vs DOW vs Standard Clustered FFS (and vs LFS)? > The Kirk softupdate stuff is as fast or (much) faster than our -async. -- John | Never try to teach a pig to sing, dyson@freebsd.org | it just makes you look stupid, jdyson@nc.com | and it irritates the pig. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Feb 10 19:00:10 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id TAA12612 for freebsd-fs-outgoing; Tue, 10 Feb 1998 19:00:10 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from smtp04.primenet.com (smtp04.primenet.com [206.165.6.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id TAA12588 for ; Tue, 10 Feb 1998 19:00:06 -0800 (PST) (envelope-from tlambert@usr04.primenet.com) Received: (from daemon@localhost) by smtp04.primenet.com (8.8.8/8.8.8) id UAA00221; Tue, 10 Feb 1998 20:00:05 -0700 (MST) Received: from usr04.primenet.com(206.165.6.204) via SMTP by smtp04.primenet.com, id smtpd000204; Tue Feb 10 20:00:04 1998 Received: (from tlambert@localhost) by usr04.primenet.com (8.8.5/8.8.5) id UAA29595; Tue, 10 Feb 1998 20:00:01 -0700 (MST) From: Terry Lambert Message-Id: <199802110300.UAA29595@usr04.primenet.com> Subject: Re: RAID support To: gibbs@plutotech.com (Justin T. Gibbs) Date: Wed, 11 Feb 1998 03:00:00 +0000 (GMT) Cc: tlambert@primenet.com, HARAWAT.IN.oracle.com.ofcmail@in.oracle.com, freebsd-fs@FreeBSD.ORG In-Reply-To: <199802102012.NAA27515@pluto.plutotech.com> from "Justin T. Gibbs" at Feb 10, 98 01:09:56 pm X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org > >Yes. I posted patchees back to the maintainers (and the -current list) > >for the RAIDFrame code. There are also several RAID SCSI controllers > >that have drivers (which are not integrated into the main FreeBSD > >code base, including one for the Compaq controller). > > There are only two controllers that I am aware of that have driver > support: the dpt, and the EISA Compaq controller. The dpt driver > has been integrated. The Compaq driver has not due to some question > about the legality of distributing it in source form. OK, I stand happily corrected. 8-). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Wed Feb 11 00:48:09 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA17405 for freebsd-fs-outgoing; Wed, 11 Feb 1998 00:48:09 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from smtp02.primenet.com (smtp02.primenet.com [206.165.6.132]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id AAA17395; Wed, 11 Feb 1998 00:48:06 -0800 (PST) (envelope-from tlambert@usr09.primenet.com) Received: (from daemon@localhost) by smtp02.primenet.com (8.8.8/8.8.8) id BAA19394; Wed, 11 Feb 1998 01:48:06 -0700 (MST) Received: from usr09.primenet.com(206.165.6.209) via SMTP by smtp02.primenet.com, id smtpd019380; Wed Feb 11 01:48:00 1998 Received: (from tlambert@localhost) by usr09.primenet.com (8.8.5/8.8.5) id BAA22120; Wed, 11 Feb 1998 01:47:58 -0700 (MST) From: Terry Lambert Message-Id: <199802110847.BAA22120@usr09.primenet.com> Subject: PATCH.D: Kernel changes for NFS client and server locking To: fs@FreeBSD.ORG, current@FreeBSD.ORG Date: Wed, 11 Feb 1998 08:47:58 +0000 (GMT) X-Mailer: ELM [version 2.4 PL25] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org This is the decription of two patches, to follow immeditely. These two patches are being submitted for review by a wider audioence to determine of they damage advisory locking semantics or other system functionality. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. ============================================================================== Description: The following two patches implement the majority of the necessary kernel code for supporting NFS client and server locking. I recommend that other *BSD's apply this against a -current FreeBSD tree, and then use that information to apply them against their own. FreeBSD has diverged greatly in the files these patches touch, despite claims to the contrary. Purpose: 1) Resubmit patches first submitted ~3 years ago. These patches are virtually unchanged, but have not been committed because they previously included bug fixes to other portions of the kernel that people (apparently) didn't want fixed. 2) Provide a small enough subset of patches at one time; the purpose of this is five-fold: a) It allows someone who has not spent their career (as I have) working on FS code to understand the nature of the changes, without me having to spend an equal amount of time as I have spent educating myself to educate them (this code is already ~3 years old; my research has progressed significantly beyond this point already). b) It increases the likelihood that they will be committed. c) It allows me to not have to merge ~1500 of my over 30,000 lines of diffs each time I update my kernel. d) It puts a damper on any forward progress I could make if all of my patches were committed, while I seperate this stuff out from my othe changes. e) It stretches out the time it's going to take to get them all committed, and generally slows down my research, since it increases the overall commit latency I must put up with. Admittedly, these last two are of questionable benefit to anyone but a wart on the tail of progress. 8-|. 3) Get feedback on whether or not this actually breaks anything for someone other than myself. It doesn't break things for me; I would be interested in the ruslts of anyone else's attempts at running the POSIX validation suite before and after these patches. 4) Provide a base from which Andrew and Doug can actually finish the implementation of rpc.lockd and the NFS client locking code, which will hopefully result in the first free and working NFS locking implementation. Jordan was going to finish this code for Kirk's class, but never did. 5) Various people at BSDI have been pestering me for these, off and on, for some time. 6) It might be nice if, for once, FreeBSD actually had something before Linux, for a change. Specific changes are: 1) Add preliminary NFS server locking support; the following fcntl() commands are added: F_RGETLK F_RSETLK F_RSETLKW F_CNVT F_UNLKSYS These are identical to the same commands in SunOS 4.1.3 2) Move VOP_ADVLOCK to a veto basis This is necessary for NFS client locking support. NFS client locking proxies local requests on the caller's behalf. 3) Add preliminary NFS client locking support. This consists of changing nfs_advlock to operate a veto basis. The ufs_advlock function is also changed so it will continue to function. 4) Add preliminary advisory locking to all filesystems. Before this patch, advisory locking was not supported on any FS other than a UFS derived FS and on NFS. 5) Seperate lock assertion from lock coelescence (see previous discussion over the past several days on the -current list). 6) Move the advisory lock list to hang off the in core vnode instead of the in core inode. This is consistent with the VOP_FINALVP discussion, and was necessary to support the veto-based VOP_ADVLOCK interface. 7) Modify vn_advlock() parameters and add vn_unlksys() support. The first is for seperation of lock commits for NFS client locking support, the second is for NFS server locking support. 8) Addition of lf_commit(). This allows a provisional lock to be committed (coelesced) or aborted (deasserted prior to coelesce). 10) Revision of lock debug dumps for use in kernel debugger. 11) Addition of a comment to union_vnops.c to note what is really taking place (necessary information for agregation support at a later date). 12) Removal of n_lockf from nfsnode. This is no longer necessary; lock lists are off the vnode now. 13) Removal of i_lockf from inode. This is no longer necessary; lock lists are off the vnode now. 14) Binary backward compatability is assured through the use of an oflock structure. Binary forward compatability is assured with an l_pad reservation of 4 longwords. 15) Copious comments 16) The fcntl.2 man page is updated to reflect these changes. ============================================================================== To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Wed Feb 11 07:07:00 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA17901 for freebsd-fs-outgoing; Wed, 11 Feb 1998 07:07:00 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from inet-user-gw-1.us.oracle.com (inet-user-gw-1.us.oracle.com [192.86.155.82]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA17891 for ; Wed, 11 Feb 1998 07:06:56 -0800 (PST) (envelope-from HARAWAT.IN.oracle.com.ofcmail@in.oracle.com) Received: from dwarpal.in.oracle.com (dwarpal.in.oracle.com [152.69.176.11]) by inet-user-gw-1.us.oracle.com (8.8.5/8.8.5) with SMTP id HAA24099 for ; Wed, 11 Feb 1998 07:04:33 -0800 (PST) Received: by dwarpal.in.oracle.com (8.6.13/37.8) id KAA12979; Wed, 11 Feb 1998 10:38:56 -0500 Message-Id: <199802111538.KAA12979@dwarpal.in.oracle.com> Date: 11 Feb 98 18:23:08 +0330 From: "HARAWAT.IN.ORACLE.COM" To: freebsd-fs@FreeBSD.ORG Subject: aio support in FreeBSD3.0 MIME-Version: 1.0 X-Mailer: Oracle InterOffice (version 4.0.5.1.55) Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I was trying to use aio calls of freebsd3.0. It took a lots of time ( almost equal the time it takes for a normal read to complete) to register aio_read request, any ideas what could be the reason for it . Harish ----------------------------------------------------------------------------- Harish Singh Rawat E-Mail : harawat@in.oracle.com Oracle Software India. Ltd. Voice : +91 (80) 228 3004/5 Indian Development Center ( IDC ) Fax : +91 (80) 225 - 6207/1591 Sona Towers No 71 Miller Road Bangalore-52 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Wed Feb 11 07:19:54 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id HAA19780 for freebsd-fs-outgoing; Wed, 11 Feb 1998 07:19:54 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from sasami.jurai.net (winter@sasami.jurai.net [207.31.78.80]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id HAA19763 for ; Wed, 11 Feb 1998 07:19:45 -0800 (PST) (envelope-from winter@jurai.net) Received: from localhost (winter@localhost) by sasami.jurai.net (8.8.8/8.8.7) with SMTP id KAA16285 for ; Wed, 11 Feb 1998 10:19:41 -0500 (EST) Date: Wed, 11 Feb 1998 10:19:40 -0500 (EST) From: "Matthew N. Dodd" To: freebsd-fs@FreeBSD.ORG Subject: NO*SYMLINK mount options? Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org What ever happened to the NO*SYMLINK mount options? Why weren't they committed? I agree that such measures are not the correct solution for buggy programs but the mount options in question -are- useful nontheless. /* Matthew N. Dodd | A memory retaining a love you had for life winter@jurai.net | As cruel as it seems nothing ever seems to http://www.jurai.net/~winter | go right - FLA M 3.1:53 */ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Thu Feb 12 02:13:04 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id CAA07681 for freebsd-fs-outgoing; Thu, 12 Feb 1998 02:13:04 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from mail.cs.tu-berlin.de (root@mail.cs.tu-berlin.de [130.149.17.13]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id CAA07676 for ; Thu, 12 Feb 1998 02:13:02 -0800 (PST) (envelope-from wosch@cs.tu-berlin.de) Received: from caramba.cs.tu-berlin.de (wosch@caramba.cs.tu-berlin.de [130.149.17.12]) by mail.cs.tu-berlin.de (8.8.6/8.8.8) with ESMTP id LAA22465; Thu, 12 Feb 1998 11:02:41 +0100 (MET) Received: (from wosch@localhost) by caramba.cs.tu-berlin.de (8.8.7/8.8.8) id LAA26545; Thu, 12 Feb 1998 11:02:27 +0100 (MET) Message-ID: <19980212110227.12668@caramba.cs.tu-berlin.de> Date: Thu, 12 Feb 1998 11:02:27 +0100 From: Wolfram Schneider To: "Matthew N. Dodd" , freebsd-fs@FreeBSD.ORG Subject: Re: NO*SYMLINK mount options? References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: ; from Matthew N. Dodd on Wed, Feb 11, 1998 at 10:19:40AM -0500 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On 1998-02-11 10:19:40 -0500, Matthew N. Dodd wrote: > What ever happened to the NO*SYMLINK mount options? Why weren't they > committed? Lack of time. > I agree that such measures are not the correct solution for > buggy programs but the mount options in question -are- useful nontheless. -- Wolfram Schneider http://www.freebsd.org/~wosch/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 17:20:32 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA26706 for freebsd-fs-outgoing; Fri, 13 Feb 1998 17:20:32 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA26699 for ; Fri, 13 Feb 1998 17:20:28 -0800 (PST) (envelope-from michaelh@cet.co.jp) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.8/CET-v2.2) with SMTP id BAA11265 for ; Sat, 14 Feb 1998 01:19:49 GMT Date: Sat, 14 Feb 1998 10:19:49 +0900 (JST) From: Michael Hancock To: freebsd-fs@FreeBSD.ORG Subject: ccd and ftd In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org I'm looking for a fault tolerant solution. I know Greg is working on a ftd driver. In the meantime I noticed that ccd supports mirroring. I'd like to mirror two arrays of 3 to 4 disk each. How well does this work? What happens when a disk goes bad? Mike Hancock -- michaelh@cet.co.jp http://www.cet.co.jp CET Inc., Daiichi Kasuya BLDG 8F, 2-5-12 Higashi Shinbashi, Minato-ku, Tokyo 105 Japan Tel: +81-3-3437-1761 Fax: +81-3-3437-1766 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 17:33:16 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id RAA28922 for freebsd-fs-outgoing; Fri, 13 Feb 1998 17:33:16 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id RAA28914 for ; Fri, 13 Feb 1998 17:33:12 -0800 (PST) (envelope-from grog@lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.8.7/8.8.5) with ESMTP id MAA23049; Sat, 14 Feb 1998 12:02:48 +1030 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.8.8/8.8.7) id MAA05927; Sat, 14 Feb 1998 12:02:47 +1030 (CST) (envelope-from grog) Message-ID: <19980214120247.43100@freebie.lemis.com> Date: Sat, 14 Feb 1998 12:02:47 +1030 From: Greg Lehey To: Michael Hancock , freebsd-fs@FreeBSD.ORG Subject: Re: ccd and ftd References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: ; from Michael Hancock on Sat, Feb 14, 1998 at 10:19:49AM +0900 WWW-Home-Page: http://www.lemis.com/~grog Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 14 February 1998 at 10:19:49 +0900, Michael Hancock wrote: > I'm looking for a fault tolerant solution. I know Greg is working on a > ftd driver. > > In the meantime I noticed that ccd supports mirroring. I'd like to mirror > two arrays of 3 to 4 disk each. How well does this work? Do you mean more than one mirror? I don't believe that works. Certainly I can't see how it can from reading the code. > What happens when a disk goes bad? The worst possible scenario. Your whole disk set goes down, and to get it up again, you have to reconfigure the set without the defective drive. That's one of the problems that my driver addresses. Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 20:39:41 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id UAA25673 for freebsd-fs-outgoing; Fri, 13 Feb 1998 20:39:41 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA25657 for ; Fri, 13 Feb 1998 20:39:36 -0800 (PST) (envelope-from michaelh@cet.co.jp) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.8/CET-v2.2) with SMTP id EAA12433; Sat, 14 Feb 1998 04:38:47 GMT Date: Sat, 14 Feb 1998 13:38:47 +0900 (JST) From: Michael Hancock To: Greg Lehey cc: freebsd-fs@FreeBSD.ORG Subject: Re: ccd and ftd In-Reply-To: <19980214120247.43100@freebie.lemis.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 14 Feb 1998, Greg Lehey wrote: > On Sat, 14 February 1998 at 10:19:49 +0900, Michael Hancock wrote: > > I'm looking for a fault tolerant solution. I know Greg is working on a > > ftd driver. > > > > In the meantime I noticed that ccd supports mirroring. I'd like to mirror > > two arrays of 3 to 4 disk each. How well does this work? > > Do you mean more than one mirror? I don't believe that works. > Certainly I can't see how it can from reading the code. The man pages says you can mirror any even number of disks. What I'd like to do is this: 2940 - disk1 - disk2 - disk3 - disk4 and mirror/duplex the whole array with another identical array. 2940 - disk5 - disk6 - disk7 - disk8 Interestingly, man ccdconfig talks about CCDF_MIRROR and an unimplemented flag CCDF_PARITY. Although, it looks like they have hooks for more fault-tolerant features in ccd I agree with your effort to make a separate fault-tolerant driver and keeping the ccd a light-weight, primarily for striping-only driver. Are you going to be able to share some code with it? Regards, Mike Hancock To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 20:58:06 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id UAA27017 for freebsd-fs-outgoing; Fri, 13 Feb 1998 20:58:06 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id UAA27011 for ; Fri, 13 Feb 1998 20:58:02 -0800 (PST) (envelope-from grog@lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.8.7/8.8.5) with ESMTP id PAA23305; Sat, 14 Feb 1998 15:27:53 +1030 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.8.8/8.8.7) id PAA03915; Sat, 14 Feb 1998 15:27:52 +1030 (CST) (envelope-from grog) Message-ID: <19980214152751.60811@freebie.lemis.com> Date: Sat, 14 Feb 1998 15:27:51 +1030 From: Greg Lehey To: Michael Hancock Cc: freebsd-fs@FreeBSD.ORG Subject: Re: ccd and ftd References: <19980214120247.43100@freebie.lemis.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: ; from Michael Hancock on Sat, Feb 14, 1998 at 01:38:47PM +0900 WWW-Home-Page: http://www.lemis.com/~grog Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 14 February 1998 at 13:38:47 +0900, Michael Hancock wrote: > On Sat, 14 Feb 1998, Greg Lehey wrote: > >> On Sat, 14 February 1998 at 10:19:49 +0900, Michael Hancock wrote: >>> I'm looking for a fault tolerant solution. I know Greg is working on a >>> ftd driver. >>> >>> In the meantime I noticed that ccd supports mirroring. I'd like to mirror >>> two arrays of 3 to 4 disk each. How well does this work? >> >> Do you mean more than one mirror? I don't believe that works. >> Certainly I can't see how it can from reading the code. > > The man pages says you can mirror any even number of disks. What I'd like > to do is this: > > 2940 - disk1 - disk2 - disk3 - disk4 > > and mirror/duplex the whole array with another identical array. > > 2940 - disk5 - disk6 - disk7 - disk8 > > Interestingly, man ccdconfig talks about CCDF_MIRROR and an unimplemented > flag CCDF_PARITY. Maybe we're talking at cross purposes here. The current version of ccd supports two mappings: 1. Concatenation (in other words, making one large space out of several smaller ones), which gives it the name. To concatenate, you don't use any flags. 2. Mirroring (redundantly storing the same data on two different spaces). To mirror, you specify the flag CCDF_MIRROR. I haven't looked at ccdconfig(8), but I suspect that if you specify more than one pair of disks on a CCDF_MIRROR line, it will concatenate pairs of mirrors. In addition, you can have striping, where the mapping is non-linear. Every certain number of bytes (stripe size), the mapping proceeds to the next component of the array, thus spreading the load more evenly. I have some PostScript documentation of the comparisons under development if anybody's interested. The CCDF_PARITY flag was for Raid 5, but as you say, it hasn't been implemented. > Although, it looks like they have hooks for more fault-tolerant > features in ccd I agree with your effort to make a separate > fault-tolerant driver and keeping the ccd a light-weight, primarily > for striping-only driver. Are you going to be able to share some > code with it? I'm a bit disappointed with CCD. I had originally planned just to extend it, but I find it too primitive. It's not just "light weight"--it's deficient. It handles faults very badly, and it offers multiple possibilities for shooting yourself in the foot. In addition, I don't think its performance is better than vinum (mine) is going to be, though that remains to be seen. About the only advantage is that it is significantly smaller, but I don't expect the typical user of either driver to be concerned about another 20 kB of memory. As a result of these problems, vinum is effectively a rewrite, though I've been looking very carefully at CCD for implementation ideas. Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 21:41:00 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA01217 for freebsd-fs-outgoing; Fri, 13 Feb 1998 21:41:00 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA01197 for ; Fri, 13 Feb 1998 21:40:56 -0800 (PST) (envelope-from michaelh@cet.co.jp) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.8/CET-v2.2) with SMTP id FAA12760; Sat, 14 Feb 1998 05:40:02 GMT Date: Sat, 14 Feb 1998 14:40:02 +0900 (JST) From: Michael Hancock To: Greg Lehey cc: freebsd-fs@FreeBSD.ORG Subject: Re: ccd and ftd In-Reply-To: <19980214152751.60811@freebie.lemis.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Another reference you might want to use is the WinNT Server Resource Kit 4.0 - Server Resource Guide. It describes the disk partition table structures and signatures used by NT FtDisk driver to track info about disks in a volume. What does vinum mean? I hope you end up making the device file something Unix-like and easy to remember like ftd. Regards, Mike On Sat, 14 Feb 1998, Greg Lehey wrote: > I'm a bit disappointed with CCD. I had originally planned just to > extend it, but I find it too primitive. It's not just "light > weight"--it's deficient. It handles faults very badly, and it offers > multiple possibilities for shooting yourself in the foot. In > addition, I don't think its performance is better than vinum (mine) is > going to be, though that remains to be seen. About the only advantage > is that it is significantly smaller, but I don't expect the typical > user of either driver to be concerned about another 20 kB of memory. > > As a result of these problems, vinum is effectively a rewrite, though > I've been looking very carefully at CCD for implementation ideas. > > Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 21:41:58 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA01492 for freebsd-fs-outgoing; Fri, 13 Feb 1998 21:41:58 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from MindBender.serv.net (mindbender.serv.net [205.153.153.98]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA01445 for ; Fri, 13 Feb 1998 21:41:52 -0800 (PST) (envelope-from mvanloon@MindBender.serv.net) From: mvanloon@MindBender.serv.net Received: from lobotomy.HeadCandy.com (Lobotomy.HeadCandy.com [198.232.197.130]) by MindBender.serv.net (8.8.8/8.7.3) with ESMTP id VAA24710; Fri, 13 Feb 1998 21:41:25 -0800 (PST) Received: by Lobotomy.HeadCandy.com with Internet Mail Service (5.5.1960.3) id <1ZAPYTS8>; Fri, 13 Feb 1998 21:40:28 -0800 Message-ID: <226EB39D3C91D111B2F900A0C9054383294C@Lobotomy.HeadCandy.com> To: grog@lemis.com, michaelh@cet.co.jp Cc: freebsd-fs@FreeBSD.ORG Subject: RE: ccd and ftd Date: Fri, 13 Feb 1998 21:40:19 -0800 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.1960.3) Content-Type: text/plain Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org You're writing a better "software-raid" driver? Here's what I want to do -- let me know if you have this in mind. I like fault-tolerance, but I don't like to take performance hits for it. I'd like to take a set of drives and mirror pairs of them. Then, I'd like to stripe on top of those mirror sets, without parity. This would give you the ultimate in performance, especially for reads, and redundant fault-tolerance, at the same time. > -----Original Message----- > From: Greg Lehey [SMTP:grog@lemis.com] > Sent: Friday, February 13, 1998 8:58 PM > To: Michael Hancock > Cc: freebsd-fs@FreeBSD.ORG > Subject: Re: ccd and ftd > > On Sat, 14 February 1998 at 13:38:47 +0900, Michael Hancock wrote: > > On Sat, 14 Feb 1998, Greg Lehey wrote: > > > >> On Sat, 14 February 1998 at 10:19:49 +0900, Michael Hancock wrote: > >>> I'm looking for a fault tolerant solution. I know Greg is working > on a > >>> ftd driver. > >>> > >>> In the meantime I noticed that ccd supports mirroring. I'd like > to mirror > >>> two arrays of 3 to 4 disk each. How well does this work? > >> > >> Do you mean more than one mirror? I don't believe that works. > >> Certainly I can't see how it can from reading the code. > > > > The man pages says you can mirror any even number of disks. What > I'd like > > to do is this: > > > > 2940 - disk1 - disk2 - disk3 - disk4 > > > > and mirror/duplex the whole array with another identical array. > > > > 2940 - disk5 - disk6 - disk7 - disk8 > > > > Interestingly, man ccdconfig talks about CCDF_MIRROR and an > unimplemented > > flag CCDF_PARITY. > > Maybe we're talking at cross purposes here. The current version of > ccd supports two mappings: > > 1. Concatenation (in other words, making one large space out of > several smaller ones), which gives it the name. To concatenate, > you don't use any flags. > > 2. Mirroring (redundantly storing the same data on two different > spaces). To mirror, you specify the flag CCDF_MIRROR. > > I haven't looked at ccdconfig(8), but I suspect that if you specify > more than one pair of disks on a CCDF_MIRROR line, it will concatenate > pairs of mirrors. > > In addition, you can have striping, where the mapping is non-linear. > Every certain number of bytes (stripe size), the mapping proceeds to > the next component of the array, thus spreading the load more evenly. > I have some PostScript documentation of the comparisons under > development if anybody's interested. > > The CCDF_PARITY flag was for Raid 5, but as you say, it hasn't been > implemented. > > > Although, it looks like they have hooks for more fault-tolerant > > features in ccd I agree with your effort to make a separate > > fault-tolerant driver and keeping the ccd a light-weight, primarily > > for striping-only driver. Are you going to be able to share some > > code with it? > > I'm a bit disappointed with CCD. I had originally planned just to > extend it, but I find it too primitive. It's not just "light > weight"--it's deficient. It handles faults very badly, and it offers > multiple possibilities for shooting yourself in the foot. In > addition, I don't think its performance is better than vinum (mine) is > going to be, though that remains to be seen. About the only advantage > is that it is significantly smaller, but I don't expect the typical > user of either driver to be concerned about another 20 kB of memory. > > As a result of these problems, vinum is effectively a rewrite, though > I've been looking very carefully at CCD for implementation ideas. > > Greg > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-fs" in the body of the message To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 22:18:16 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id WAA05374 for freebsd-fs-outgoing; Fri, 13 Feb 1998 22:18:16 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id WAA05368 for ; Fri, 13 Feb 1998 22:18:14 -0800 (PST) (envelope-from michaelh@cet.co.jp) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.8/CET-v2.2) with SMTP id GAA12947; Sat, 14 Feb 1998 06:17:27 GMT Date: Sat, 14 Feb 1998 15:17:27 +0900 (JST) From: Michael Hancock To: Greg Lehey cc: freebsd-fs@FreeBSD.ORG Subject: Re: ccd and ftd In-Reply-To: <19980214152751.60811@freebie.lemis.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 14 Feb 1998, Greg Lehey wrote: > > 2940 - disk1 - disk2 - disk3 - disk4 > > > > and mirror/duplex the whole array with another identical array. > > > > 2940 - disk5 - disk6 - disk7 - disk8 > > > Maybe we're talking at cross purposes here. The current version of Yes, I'd like to mirror 2 arrays and stripe them. Presumably, your driver would allow you to do the following: 1) Mirroring 2) Concatenating and mirroring 3) Striping and mirroring (expensive, fast, and safe) 4) Striping with parity (RAID5) (cheaper, pretty fast, and safe) I don't think anything else is required. For anything less than fault-tolerant you'd just use ccd. Mike To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 23:06:08 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA09660 for freebsd-fs-outgoing; Fri, 13 Feb 1998 23:06:08 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA09649 for ; Fri, 13 Feb 1998 23:06:02 -0800 (PST) (envelope-from grog@lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.8.7/8.8.5) with ESMTP id RAA23405; Sat, 14 Feb 1998 17:35:50 +1030 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.8.8/8.8.7) id RAA03015; Sat, 14 Feb 1998 17:35:50 +1030 (CST) (envelope-from grog) Message-ID: <19980214173549.11419@freebie.lemis.com> Date: Sat, 14 Feb 1998 17:35:49 +1030 From: Greg Lehey To: Michael Hancock Cc: freebsd-fs@FreeBSD.ORG Subject: Re: ccd and ftd References: <19980214152751.60811@freebie.lemis.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: ; from Michael Hancock on Sat, Feb 14, 1998 at 02:40:02PM +0900 WWW-Home-Page: http://www.lemis.com/~grog Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 14 February 1998 at 14:40:02 +0900, Michael Hancock wrote: > On Sat, 14 Feb 1998, Greg Lehey wrote: > >> I'm a bit disappointed with CCD. I had originally planned just to >> extend it, but I find it too primitive. It's not just "light >> weight"--it's deficient. It handles faults very badly, and it offers >> multiple possibilities for shooting yourself in the foot. In >> addition, I don't think its performance is better than vinum (mine) is >> going to be, though that remains to be seen. About the only advantage >> is that it is significantly smaller, but I don't expect the typical >> user of either driver to be concerned about another 20 kB of memory. >> >> As a result of these problems, vinum is effectively a rewrite, though >> I've been looking very carefully at CCD for implementation ideas.> > > Another reference you might want to use is the WinNT Server Resource Kit > 4.0 - Server Resource Guide. It describes the disk partition table > structures and signatures used by NT FtDisk driver to track info about > disks in a volume. Is this different from the current Microsoft partition table? My current intention is to add a supplementary disk label (thus solving the ccd problem of not knowing what's on the disk). But it wouldn't be difficult to change that. > What does vinum mean? It's Latin for wine. You may recognize the quotation "In vino veritas" :-) It leans somewhat on the Veritas Volume Manager. > I hope you end up making the device file something Unix-like and > easy to remember like ftd. Better than that. You get to name the device files yourself. They're on the disk, so when the vinum LKM is started, it finds the volumes (wherever they may be) and creates the device files. Things will, of course, improve with devfs. Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 23:19:00 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA11224 for freebsd-fs-outgoing; Fri, 13 Feb 1998 23:19:00 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA11219 for ; Fri, 13 Feb 1998 23:18:53 -0800 (PST) (envelope-from grog@lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.8.7/8.8.5) with ESMTP id RAA23416; Sat, 14 Feb 1998 17:48:50 +1030 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.8.8/8.8.7) id RAA03053; Sat, 14 Feb 1998 17:48:45 +1030 (CST) (envelope-from grog) Message-ID: <19980214174844.41352@freebie.lemis.com> Date: Sat, 14 Feb 1998 17:48:44 +1030 From: Greg Lehey To: mvanloon@mindbender.serv.net, michaelh@cet.co.jp Cc: freebsd-fs@FreeBSD.ORG Subject: Re: ccd and ftd References: <226EB39D3C91D111B2F900A0C9054383294C@Lobotomy.HeadCandy.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: <226EB39D3C91D111B2F900A0C9054383294C@Lobotomy.HeadCandy.com>; from mvanloon@mindbender.serv.net on Fri, Feb 13, 1998 at 09:40:19PM -0800 WWW-Home-Page: http://www.lemis.com/~grog Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Fri, 13 February 1998 at 21:40:19 -0800, mvanloon@mindbender.serv.net wrote: >> -----Original Message----- >> From: Greg Lehey [SMTP:grog@lemis.com] >> Sent: Friday, February 13, 1998 8:58 PM >> To: Michael Hancock >> Cc: freebsd-fs@FreeBSD.ORG >> Subject: Re: ccd and ftd >> >> On Sat, 14 February 1998 at 13:38:47 +0900, Michael Hancock wrote: >>> On Sat, 14 Feb 1998, Greg Lehey wrote: >>> >>>> On Sat, 14 February 1998 at 10:19:49 +0900, Michael Hancock wrote: >>>>> I'm looking for a fault tolerant solution. I know Greg is working >> on a >>>>> ftd driver. >>>>> >>>>> In the meantime I noticed that ccd supports mirroring. I'd like >> to mirror >>>>> two arrays of 3 to 4 disk each. How well does this work? >>>> >>>> Do you mean more than one mirror? I don't believe that works. >>>> Certainly I can't see how it can from reading the code. >>> >>> The man pages says you can mirror any even number of disks. What >> I'd like >>> to do is this: >>> >>> 2940 - disk1 - disk2 - disk3 - disk4 >>> >>> and mirror/duplex the whole array with another identical array. >>> >>> 2940 - disk5 - disk6 - disk7 - disk8 >>> >>> Interestingly, man ccdconfig talks about CCDF_MIRROR and an >> unimplemented >>> flag CCDF_PARITY. >> >> Maybe we're talking at cross purposes here. The current version of >> ccd supports two mappings: >> >> 1. Concatenation (in other words, making one large space out of >> several smaller ones), which gives it the name. To concatenate, >> you don't use any flags. >> >> 2. Mirroring (redundantly storing the same data on two different >> spaces). To mirror, you specify the flag CCDF_MIRROR. >> >> I haven't looked at ccdconfig(8), but I suspect that if you specify >> more than one pair of disks on a CCDF_MIRROR line, it will concatenate >> pairs of mirrors. >> >> In addition, you can have striping, where the mapping is non-linear. >> Every certain number of bytes (stripe size), the mapping proceeds to >> the next component of the array, thus spreading the load more evenly. >> I have some PostScript documentation of the comparisons under >> development if anybody's interested. >> >> The CCDF_PARITY flag was for Raid 5, but as you say, it hasn't been >> implemented. >> >>> Although, it looks like they have hooks for more fault-tolerant >>> features in ccd I agree with your effort to make a separate >>> fault-tolerant driver and keeping the ccd a light-weight, primarily >>> for striping-only driver. Are you going to be able to share some >>> code with it? >> >> I'm a bit disappointed with CCD. I had originally planned just to >> extend it, but I find it too primitive. It's not just "light >> weight"--it's deficient. It handles faults very badly, and it offers >> multiple possibilities for shooting yourself in the foot. In >> addition, I don't think its performance is better than vinum (mine) is >> going to be, though that remains to be seen. About the only advantage >> is that it is significantly smaller, but I don't expect the typical >> user of either driver to be concerned about another 20 kB of memory. >> >> As a result of these problems, vinum is effectively a rewrite, though >> I've been looking very carefully at CCD for implementation ideas. > > You're writing a better "software-raid" driver? Something like that. > Here's what I want to do -- let me know if you have this in mind. > > I like fault-tolerance, but I don't like to take performance hits for > it. I'd like to take a set of drives and mirror pairs of them. Then, > I'd like to stripe on top of those mirror sets, without parity. > > This would give you the ultimate in performance, especially for reads, > and redundant fault-tolerance, at the same time. Yes, it will be able to do that. Specifically (those of you who know Veritas will recognize this): 1. Each partitions is split up into an arbitary number of subdisks, contiguous sections of disk space. For best performance, you would take one subdisk per volume. 2. The subdisks are combined into plexes. Within a plex, each subdisk represents a unique part of the address space (i.e. there is no redundancy). There is no requirement for the address space of a plex to be contiguous. Plexes can be organized in three ways: sequentially (i.e. each subdisk represents a contiguous part of the plex address space), striped (in other words, the address space is patched together out of fixed-length pieces sequentially allocated from each subdisk), or RAID 5 (like striping, except that in each group of stripes one subdisk is reserved for a parity stripe). I realise that this description isn't the utmost in clarity. If somebody can help me phrase it better, I'd be grateful. 3. Up to 8 plexes can be combined to form a volume. The volume corresponds to a normal disk device. Each plex replicates the same address space (i.e. they provide mirroring). There is a requirement that the logical sum of the plexes provide storage for the complete address space of the volume. Mapping this to your requirements, you'd get the striping by taking striped plexes, and the redundancy from multiple plexes per volume. There are a couple of other advantages to this method: 1. You can move data from one disk to another on line without loss of fault tolerance by dynamically adding a new plex and then removing the old one. 2. You can make consistent backups of a volume by adding a plex, and then detaching it at a specific time. Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Fri Feb 13 23:19:56 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id XAA11339 for freebsd-fs-outgoing; Fri, 13 Feb 1998 23:19:56 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from allegro.lemis.com (allegro.lemis.com [192.109.197.134]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id XAA11333 for ; Fri, 13 Feb 1998 23:19:52 -0800 (PST) (envelope-from grog@lemis.com) Received: from freebie.lemis.com (freebie.lemis.com [192.109.197.137]) by allegro.lemis.com (8.8.7/8.8.5) with ESMTP id RAA23420; Sat, 14 Feb 1998 17:49:50 +1030 (CST) Received: (from grog@localhost) by freebie.lemis.com (8.8.8/8.8.7) id RAA03066; Sat, 14 Feb 1998 17:49:49 +1030 (CST) (envelope-from grog) Message-ID: <19980214174949.03637@freebie.lemis.com> Date: Sat, 14 Feb 1998 17:49:49 +1030 From: Greg Lehey To: Michael Hancock Cc: freebsd-fs@FreeBSD.ORG Subject: Re: ccd and ftd References: <19980214152751.60811@freebie.lemis.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.89i In-Reply-To: ; from Michael Hancock on Sat, Feb 14, 1998 at 03:17:27PM +0900 WWW-Home-Page: http://www.lemis.com/~grog Organization: LEMIS, PO Box 460, Echunga SA 5153, Australia Phone: +61-8-8388-8286 Fax: +61-8-8388-8725 Mobile: +61-41-739-7062 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 14 February 1998 at 15:17:27 +0900, Michael Hancock wrote: > On Sat, 14 Feb 1998, Greg Lehey wrote: > >>> 2940 - disk1 - disk2 - disk3 - disk4 >>> >>> and mirror/duplex the whole array with another identical array. >>> >>> 2940 - disk5 - disk6 - disk7 - disk8 >>> >> Maybe we're talking at cross purposes here. The current version of > > Yes, I'd like to mirror 2 arrays and stripe them. Presumably, your driver > would allow you to do the following: > > 1) Mirroring > > 2) Concatenating and mirroring > > 3) Striping and mirroring (expensive, fast, and safe) > > 4) Striping with parity (RAID5) (cheaper, pretty fast, and safe) Yes. See my last message. > I don't think anything else is required. Neither do I, otherwise I would have done it :-) > For anything less than fault-tolerant you'd just use ccd. Why? Greg To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sat Feb 14 00:55:45 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id AAA20578 for freebsd-fs-outgoing; Sat, 14 Feb 1998 00:55:45 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from parkplace.cet.co.jp (parkplace.cet.co.jp [202.32.64.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id AAA20542 for ; Sat, 14 Feb 1998 00:55:38 -0800 (PST) (envelope-from michaelh@cet.co.jp) Received: from localhost (michaelh@localhost) by parkplace.cet.co.jp (8.8.8/CET-v2.2) with SMTP id IAA13699; Sat, 14 Feb 1998 08:54:47 GMT Date: Sat, 14 Feb 1998 17:54:47 +0900 (JST) From: Michael Hancock To: Greg Lehey cc: freebsd-fs@FreeBSD.ORG Subject: Re: ccd and ftd In-Reply-To: <19980214173549.11419@freebie.lemis.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Sat, 14 Feb 1998, Greg Lehey wrote: > > Another reference you might want to use is the WinNT Server Resource Kit > > 4.0 - Server Resource Guide. It describes the disk partition table > > structures and signatures used by NT FtDisk driver to track info about > > disks in a volume. > > Is this different from the current Microsoft partition table? My > current intention is to add a supplementary disk label (thus solving > the ccd problem of not knowing what's on the disk). But it wouldn't > be difficult to change that. NTFS does things differently than FAT with the tables if that's what you mean. Another reason I think it'd be worthwhile looking into in case someone implements NTFS and want's to use vinum. Mike To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sat Feb 14 06:00:24 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id GAA21962 for freebsd-fs-outgoing; Sat, 14 Feb 1998 06:00:24 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from gjp.erols.com (alex-va-n008c243.moon.jic.com [206.156.18.253]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id GAA21953 for ; Sat, 14 Feb 1998 06:00:23 -0800 (PST) (envelope-from gjp@gjp.erols.com) Received: from gjp.erols.com (localhost.erols.com [127.0.0.1]) by gjp.erols.com (8.8.8/8.8.7) with ESMTP id IAA06959; Sat, 14 Feb 1998 08:58:32 -0500 (EST) (envelope-from gjp@gjp.erols.com) X-Mailer: exmh version 2.0.1 12/23/97 To: Greg Lehey cc: Michael Hancock , freebsd-fs@FreeBSD.ORG From: "Gary Palmer" Subject: Re: ccd and ftd In-reply-to: Your message of "Sat, 14 Feb 1998 12:02:47 +1030." <19980214120247.43100@freebie.lemis.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Sat, 14 Feb 1998 08:58:31 -0500 Message-ID: <6955.887464711@gjp.erols.com> Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Greg Lehey wrote in message ID <19980214120247.43100@freebie.lemis.com>: > On Sat, 14 February 1998 at 10:19:49 +0900, Michael Hancock wrote: > > What happens when a disk goes bad? > The worst possible scenario. Your whole disk set goes down, and to > get it up again, you have to reconfigure the set without the defective > drive. That's one of the problems that my driver addresses. And to make it worse, filesystem operations may fail if one half of the mirror is bad, leading to a really badly trashed filesystem once you reboot and have removed the bad mirror. This, IMHO, is bad, and why CCD mirroring should not be used without some advanced monitoring system that can break the mirror at the first sign of disk errors (possibly after trying to re-sync the mirrors to see if it was a transient error). Gary -- Gary Palmer FreeBSD Core Team Member FreeBSD: Turning PC's into workstations. See http://www.FreeBSD.ORG/ for info To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Sat Feb 14 06:10:00 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id GAA25110 for freebsd-fs-outgoing; Sat, 14 Feb 1998 06:10:00 -0800 (PST) (envelope-from owner-freebsd-fs@FreeBSD.ORG) Received: from proxy-bsb.gns.com.br (a1E+HVLkemZvjhghx2dieijpgG7hDiMr@srv1-bsb.GNS.com.br [200.239.56.1]) by hub.freebsd.org (8.8.8/8.8.8) with SMTP id GAA25085 for ; Sat, 14 Feb 1998 06:09:52 -0800 (PST) (envelope-from dcs@gns.com.br) Received: (qmail 13673 invoked by uid 1147); 14 Feb 1998 13:55:20 -0000 Received: from srv1-bsb.gns.com.br (200.239.56.1) by proxy-bsb.gns.com.br with SMTP; 14 Feb 1998 13:55:20 -0000 Received: (from mail@localhost) by srv1-bsb.gns.com.br (8.8.8/8.8.8) id LAA13669; Sat, 14 Feb 1998 11:55:20 -0200 (EDT) Received: from dl0235-bsb.gns.com.br(200.239.56.235) by srv1-bsb.gns.com.br via smap (V2.0) id xma013662; Sat, 14 Feb 98 11:54:27 -0200 Received: (from dcs@localhost) by daniel.sobral (8.8.8/8.8.6) id MAA00477; Sat, 14 Feb 1998 12:00:54 -0200 (EDT) From: "Daniel C. Sobral" Message-Id: <199802141400.MAA00477@daniel.sobral> Subject: Re: ccd and ftd In-Reply-To: <19980214173549.11419@freebie.lemis.com> from Greg Lehey at "Feb 14, 98 05:35:49 pm" To: grog@lemis.com (Greg Lehey) Date: Sat, 14 Feb 1998 12:00:54 -0200 (EDT) Disclaimer: Klaatu Barada Nikto! X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Unheedful of thy elder's warnings, Greg Lehey wrote: > > Is this different from the current Microsoft partition table? My > current intention is to add a supplementary disk label (thus solving > the ccd problem of not knowing what's on the disk). But it wouldn't > be difficult to change that. BTW, do you intend to provide support for hot spares? SNMP? Executing programs on alerts or sending e-mail? The later two are obvious. As for hot spares, they are just disks that sit waiting for one disk in any of the configured raid arrays to fail, and then take their place automatically. -- Daniel C. Sobral (8-DCS) dcs@gns.com.br Be careful when you bite into your hamburger. -- Derek Bok To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message