From owner-freebsd-fs Sun Jul 8 14: 6:56 2001 Delivered-To: freebsd-fs@freebsd.org Received: from salmon.maths.tcd.ie (salmon.maths.tcd.ie [134.226.81.11]) by hub.freebsd.org (Postfix) with SMTP id D718137B405 for ; Sun, 8 Jul 2001 14:06:52 -0700 (PDT) (envelope-from iedowse@maths.tcd.ie) Received: from walton.maths.tcd.ie by salmon.maths.tcd.ie with SMTP id ; 8 Jul 2001 22:06:52 +0100 (BST) To: freebsd-fs@FreeBSD.ORG Cc: Matt Dillon , mckusick@mckusick.com Subject: Re: CFR: UFS directory hashing In-Reply-To: Your message of "Tue, 03 Jul 2001 10:53:49 PDT." <200107031753.f63Hrnd33021@earth.backplane.com> Date: Sun, 08 Jul 2001 22:06:51 +0100 From: Ian Dowse Message-ID: <200107082206.aa68029@salmon.maths.tcd.ie> Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org In message <200107031753.f63Hrnd33021@earth.backplane.com>, Matt Dillon writes: > will help. A delayed approach to hashing the directory might > work, you could use the directory vnode or the namei cache > (associated with the directory itself) to store the statistics > or something like that. Ok, after thinking about this for a while, I have made a relatively simple change to the recycling code that seems to improve the situation considerably. The new algorithm is pretty arbitrary, but it significantly reduces the number of hash-build operations that occur when the working set is larger than the memory limit. The old algorithm was simple LRU, so a working set larger than the memory limit resulted in every lookup causing a hash build. The new approach is a kind of a hybrid between LRU and least-often-used with an adjustment that results in only a small fraction of lookups causing a hash build in the overload case. Each hash is placed on a LRU linked list as before, with new entries inserted at the MRU end of the list. Candidates for recycling are taken from the LRU end. Each hash now has a `score' that is incremented when a hash is used and decremented when the hash is a candidate for recycling. A hash is only recycled if its score reaches zero, and on access, a hash is moved one slot towards the MRU end of the list iff its score is greater than that of the next entry. The score of a new hash entry is set to 8, and scores are limited to a maximum of 64. The initial value means that in the worst-case thrashing situation, at most 1/8 of lookups result in a hash build. The least-often-used discard policy also improves the chance of holding on to a hashed directory across a large burst of accesses to other directories. It might be better to weight the score increments and decrements by the directory size, but that introduces some other problems so I'd prefer to keep it simple for now. The new code, which is at http://www.maths.tcd.ie/~iedowse/FreeBSD/dirhash.diff6 also includes a performance improvement and a bugfix to the sequential access optimisation. I haven't updated the releng_4 version, but I will soon. Ian To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Tue Jul 10 7:24:11 2001 Delivered-To: freebsd-fs@freebsd.org Received: from web10105.mail.yahoo.com (web10105.mail.yahoo.com [216.136.130.55]) by hub.freebsd.org (Postfix) with SMTP id 7E7E837B408 for ; Tue, 10 Jul 2001 07:24:06 -0700 (PDT) (envelope-from kidmamey2k@yahoo.com) Message-ID: <20010710142406.52433.qmail@web10105.mail.yahoo.com> Received: from [148.233.159.25] by web10105.mail.yahoo.com via HTTP; Tue, 10 Jul 2001 07:24:06 PDT Date: Tue, 10 Jul 2001 07:24:06 -0700 (PDT) From: Kid Mamey Subject: Problems with my FS To: freebsd-questions@FreeBSD.ORG, freebsd-fs@FreeBSD.ORG MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org I really need help related to my filesystem. During startup fsck tell me that I need to run it manually, running manually I get the following output: ad0: HARD READ ERROR blk# 8775024 status=59 error=40 cannot read: BLK 2096896 and also HARD READ ERROR blk# 8775028 status=59 error=40 the follow disk sector could not be read read: BLK 2096900 My questions are: 1.-Is there any way I can patch these blocks without formating? 2.-Does formating the filesystem repairs the damage? 3.-Right now I'm running in read-only FS mode. Is there a way I can copy my data into some other partition(changing the fs mode), say my windows partition before formating and also how can I mount the other partitions in the fstab? 4.-My cd-rom is via pcmcia, can I make it boot for installation, or booting with the installation diskettes detects it and gets all I need? thank you Gerardo Amaya __________________________________________________ Do You Yahoo!? Get personalized email addresses from Yahoo! Mail http://personal.mail.yahoo.com/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Thu Jul 12 5:33:18 2001 Delivered-To: freebsd-fs@freebsd.org Received: from platan.uvt.ro (platan.uvt.ro [193.226.14.130]) by hub.freebsd.org (Postfix) with ESMTP id EA5D837B405 for ; Thu, 12 Jul 2001 05:33:13 -0700 (PDT) (envelope-from iwtk75@platan.uvt.ro) Received: from localhost (iwtk75@localhost) by platan.uvt.ro (8.11.1/8.11.1) with ESMTP id f6CFVlg04862 for ; Thu, 12 Jul 2001 18:31:54 +0300 (EEST) (envelope-from iwtk75@platan.uvt.ro) Date: Thu, 12 Jul 2001 18:31:46 +0300 (EEST) From: Varan To: Message-ID: <20010712183052.V4798-100000@platan.uvt.ro> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org subscribe freebsd-fs To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Thu Jul 12 12: 9: 8 2001 Delivered-To: freebsd-fs@freebsd.org Received: from obsecurity.dyndns.org (adsl-63-207-60-73.dsl.lsan03.pacbell.net [63.207.60.73]) by hub.freebsd.org (Postfix) with ESMTP id 24BA337B405; Thu, 12 Jul 2001 12:09:01 -0700 (PDT) (envelope-from kris@obsecurity.org) Received: by obsecurity.dyndns.org (Postfix, from userid 1000) id 8436C66D72; Thu, 12 Jul 2001 12:08:59 -0700 (PDT) Date: Thu, 12 Jul 2001 12:08:59 -0700 From: Kris Kennaway To: Kid Mamey Cc: freebsd-questions@FreeBSD.ORG, freebsd-fs@FreeBSD.ORG Subject: Re: Problems with my FS Message-ID: <20010712120858.A6019@xor.obsecurity.org> References: <20010710142406.52433.qmail@web10105.mail.yahoo.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-md5; protocol="application/pgp-signature"; boundary="bg08WKrSYDhXBjb5" Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20010710142406.52433.qmail@web10105.mail.yahoo.com>; from kidmamey2k@yahoo.com on Tue, Jul 10, 2001 at 07:24:06AM -0700 Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org --bg08WKrSYDhXBjb5 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Tue, Jul 10, 2001 at 07:24:06AM -0700, Kid Mamey wrote: > I really need help related to my filesystem. During > startup fsck tell me that I=20 > need to run it manually, running manually I get the > following output:=20 >=20 > ad0: HARD READ ERROR blk# 8775024 status=3D59 error=3D40=20 > cannot read: BLK 2096896=20 Your disk is bad. Probably nothing you can do about it except copy off what remaining data you can, before it dies completely. Kris --bg08WKrSYDhXBjb5 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (FreeBSD) Comment: For info see http://www.gnupg.org iD8DBQE7TfXKWry0BWjoQKURAioMAKDLtpre5P7YMivs+AMlvU6lank2/wCeP87e eeTGGzTX5poUtxER0jgM9yE= =cNBf -----END PGP SIGNATURE----- --bg08WKrSYDhXBjb5-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message From owner-freebsd-fs Thu Jul 12 15:57: 6 2001 Delivered-To: freebsd-fs@freebsd.org Received: from peter3.wemm.org (c1315225-a.plstn1.sfba.home.com [24.14.150.180]) by hub.freebsd.org (Postfix) with ESMTP id 743C637B403; Thu, 12 Jul 2001 15:56:56 -0700 (PDT) (envelope-from peter@wemm.org) Received: from overcee.netplex.com.au (overcee.wemm.org [10.0.0.3]) by peter3.wemm.org (8.11.0/8.11.0) with ESMTP id f6CMuuM96285; Thu, 12 Jul 2001 15:56:56 -0700 (PDT) (envelope-from peter@wemm.org) Received: from wemm.org (localhost [127.0.0.1]) by overcee.netplex.com.au (Postfix) with ESMTP id 75AD13811; Thu, 12 Jul 2001 15:56:53 -0700 (PDT) (envelope-from peter@wemm.org) X-Mailer: exmh version 2.3.1 01/18/2001 with nmh-1.0.4 To: Kris Kennaway Cc: Kid Mamey , freebsd-questions@FreeBSD.ORG, freebsd-fs@FreeBSD.ORG Subject: Re: Problems with my FS In-Reply-To: <20010712120858.A6019@xor.obsecurity.org> Date: Thu, 12 Jul 2001 15:56:53 -0700 From: Peter Wemm Message-Id: <20010712225653.75AD13811@overcee.netplex.com.au> Sender: owner-freebsd-fs@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Kris Kennaway wrote: > > --bg08WKrSYDhXBjb5 > Content-Type: text/plain; charset=us-ascii > Content-Disposition: inline > Content-Transfer-Encoding: quoted-printable > > On Tue, Jul 10, 2001 at 07:24:06AM -0700, Kid Mamey wrote: > > I really need help related to my filesystem. During > > startup fsck tell me that I=20 > > need to run it manually, running manually I get the > > following output:=20 > >=20 > > ad0: HARD READ ERROR blk# 8775024 status=3D59 error=3D40=20 > > cannot read: BLK 2096896=20 > > Your disk is bad. Probably nothing you can do about it except copy > off what remaining data you can, before it dies completely. > > Kris .. Unless it is an IBM DTLA series. In which case it will do this at will. One "solution" is to repartition the disk so that you use the first 29 out of 30GB or so. The other "solution" is to throw it in the bin and get a different brand disk, although not many find this an acceptable option. If you dont have an IBM DTLA, then please ignore. :-) Cheers, -Peter -- Peter Wemm - peter@FreeBSD.org; peter@yahoo-inc.com; peter@netplex.com.au "All of this is for nothing if we don't go to the stars" - JMS/B5 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-fs" in the body of the message