From owner-freebsd-questions Sat Nov 27 9:16:59 1999 Delivered-To: freebsd-questions@freebsd.org Received: from mail2.gmx.net (mail2.gmx.net [194.221.183.62]) by hub.freebsd.org (Postfix) with SMTP id 45F9C14C36 for ; Sat, 27 Nov 1999 09:16:32 -0800 (PST) (envelope-from pons@gmx.li) Received: (qmail 25869 invoked by uid 0); 27 Nov 1999 17:16:21 -0000 Received: from unknown (HELO gmx.li) (212.38.131.194) by mail2.gmx.net with SMTP; 27 Nov 1999 17:16:21 -0000 Message-ID: <384010D4.635DE54D@gmx.li> Date: Sat, 27 Nov 1999 19:11:48 +0200 From: pons X-Mailer: Mozilla 4.7 [en] (Win98; I) X-Accept-Language: en,arabic MIME-Version: 1.0 Cc: freebsd-questions@freebsd.org Subject: Re: Harddisk problems References: <199911271244.NAA18495@wjpserver.cs.uni-sb.de> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Thomas Schuerger wrote: > > Hi! > > I just bought an IBM Deskstar DPTA-372730 27.3 GB UDMA/2 drive > for my FreeBSD 4.0-CURRENT system. However, I am having severe problems > with the 19 GB slice I have created. > > I am using the ATA driver with my UDMA controller on my ASUS P2B-DS > mainboard. The ATA driver keeps telling me there are "hard errors" > when reading some certain sectors on the disk. fsck currently cannot > check the slice, as it reports there are two sectors that cannot be > read (fsck does not mark the slice as being clean, so I cannot > mount the disk r/w). However, before there were once problems to > unmount the disk on reboot (and therefore fsck check was enforced), > I was able to use the slice just fine (even though the ATA driver > reported those errors in the console). The other slices on the disk > work fine so far. > > The ATA driver outputs the following in the console: > > ad0: status=51 error=84 > ad_interrupt: hard error > > fsck reports the following: > > ** /dev/rad0s1g > ** Last Mounted on /data > ** Phase 1 - Check Blocks and Sizes > > CANNOT READ: BLK 18350144 > CONTINUE? [yn] y > > THE FOLLOWING DISK SECTORS COULD NOT BE READ: 18350191, > > CANNOT READ: BLK 38274656 > CONTINUE? [yn] y > > THE FOLLOWING DISK SECTORS COULD NOT BE READ: 38274713, > ** Phase 2 - Check Pathnames > ** Phase 3 - Check Connectivity > ** Phase 4 - Check Reference Counts > ** Phase 5 - Check Cyl groups > 13387 files, 7898160 used, 11063158 free (11870 frags, 1381411 blocks, 0.1% fragmentation) > > ***** FILE SYSTEM STILL DIRTY ***** > > ***** PLEASE RERUN FSCK ***** > > I have tried formatting the whole partition with Windows and that > worked fine. Then I used newfs to reinstall UFS on it (which went > much too fast for actually formatting the disk), something like > 30 seconds with FreeBSD compared to 20 minutes with Windows (note > that using Windows I formatted the whole partition, whereas using > FreeBSD I just newfs'ed the 19 GB slice). > > Is there a better way to actually check the slice for errors or > do a "format" that does a verify on the formatted data? I'd like to > know whether the disk really has some hard errors on it or if perhaps > the filesystem or ATA driver cannot handle such big partition. I'd > also like to know which testing tools can be used to check that. > > Thanks for your answers! > > Ciao, > Thomas Schürger. http://www.menden.augustin.de > Hi i can tell you about my story, I had before few days a HDD crash (Bad Cluster) with my Maxtor IDE/AT, i resolved it by downloading a Low level format program MAXLLF.EXE from the web "The MAXLLF.EXE program is designed and intended to be used only with Maxtor IDE/AT hard drives. This program will Low Level format the hard drive and restore the media to it's original factory format." Maxtor's Low Level Format Utility can be obtained from http://www.maxtor.com/ That was from my own experience. -- pons@gmx.li http://neptune.spaceports.com/~pons/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message