Data recovery and disk repair questions and discussions related to old-fashioned SATA, SAS, SCSI, IDE, MFM hard drives - any type of storage device that has moving parts
September 19th, 2019, 23:48
Noob here, doing a data recovery on a 13-15 year old laptop hard drive, using ddrescue I quickly cloned about 8% of the hard drive, then it started clicking. Oddly enough it will click consistently (takes about 5-15 seconds to error out and allow reading another sector) if I read certain sectors but it reads reliably on any sector previously read successfully. It doesn't click when powered on or any time after unless I try reading a bad area. The drive is formatted NTFS and the whole MFT was within the 8% previously cloned so I narrowed down my data recovery to about 200MB of the remaining 36GB. After some time yolo-ing ddrescue reading one sector at a time backwards with a mapfile generated specifically for the 200MB I need to clone; I was able to squeeze out another 100MB, but now I cant seem to read any more new data even skipping around to different areas using ddrescue -i and -s options.
The drive doesn't have any ERC/TLER/CCTL so I cant reduce timeouts to get faster scrubbing with ddrescue and I already have linux's timeout, eh_timeout, io_timeout set to the lowest values. I have not checked to see if ATA readlong works yet and I probably should because the drive is old enough to have a good chance of that working correctly.
Is there anything I'm missing? maybe a well known VSC that is the equivalent of setting ERC? firmware issue? I just need the hard drive to give me 100 more megabytes.
September 20th, 2019, 4:20
upload the ddrescue log file
September 24th, 2019, 22:13
HaspHL wrote:upload the ddrescue log file
Sorry about the delay, here is the ddrescue mapfile. Please note that almost all of the 100MB of bad sectors in the logfile are from when the drive dropped off the IDE bus and are not real bad sectors.
Here is the logfile in ddrescueview, note that the repeating pattern of non-trimmed sectors spaced evenly is from me specifying a large skip value on failed reads:
Here is that same log file with my domain logfile of important sectors applied:
Powered by phpBB © phpBB Group.