|
From: Tim S. <ti...@se...> - 2014-12-17 09:49:58
|
On 17/12/14 09:03, Kimio Miyamura wrote: > Here, I got contradiction. The LBA provided by short test is bigger > than fdisk's sector size. Also, with smratctl -a command, I got > different LBA from short test result. > > Even though long test passed, short test fails always. So I am a bit > worry about it. My request is how to fix read failure in Mac OS X. If there's doubt about what the actual LBA is, then to be on the safe side, I would try and use dd to read the entire disk: dd if=/dev/disk5 of=/dev/null bs=512 on Linux I'd also use iflag=direct to stop any OS level buffering, but don't know how this is done (if at all) on OSX. You will then confirm where (if anywhere user-addressable) the error actually is. Reconfirm using dd if=/dev/disk5 of=/dev/null bs=512 skip=suspectedLBA count=1 (again adding option for direct I/O if possible). I'd then use dd with bs=512 if=/dev/zero count=1 seek=suspectedLBA to write to the correct disk device, and then confirm pending sector = 0 in smartctl output. If you can't find the sector, then maybe the drive is using something like DCO. In that case maybe best to use a Linux boot disk (something like GRML, which includes smartctl and hdparm) to proceed with minimal risk to your OS install. Tim. -- South East Open Source Solutions Limited Registered in England and Wales with company number 06134732. Registered Office: 2 Powell Gardens, Redhill, Surrey, RH1 1TQ VAT number: 900 6633 53 http://seoss.co.uk/ +44-(0)1273-808309 |