View Single Post
Old 15th August 2011
jggimi's Avatar
jggimi jggimi is offline
More noise than signal
 
Join Date: May 2008
Location: USA
Posts: 7,977
Default

If I'm reading this right, you have 0x10, or 16 "Offline Uncorrectable" sectors, and 0x12, or 18 "Current Pending Sectors."

If I understand SMART, the 16 offline uncorrectable errors are from self-tests performed by the electronics at some point. Pending sectors, on the other hand, are sectors that failed to read when requested by an OS, and those sector numbers are awaiting writes ... at which time, the hardware will assign sectors from a selection of spares and remove them from the list. Welcome to modern disk hardware, when bad sectors are pretty much outside of your control; all you can do is write over them again and have the hardware put the new data on some other physical location on the drive.

You've also had a serious heat problem, which may have been related to your power supply failure, and may have been the root cause of this media damage.

Your data, or the majority of it, should be retrievable, if the drive electronics are functioning properly, unless one of the failed sectors is critical to the filesystem. And even then, it may be possible to retrieve complete files or parts of files, using forensic tools.

----

I am baffled by the failures of userland programs like fdisk and dd. Especially since fdisk, with or without -e, was just reading sector 0.

In general, when you conduct an I/O to a disk drive, you expect one of four responses:
  1. Good results, data moved.
  2. Errors occurred. On OpenBSD, the kernel produces messages.
  3. No response at all -- on OpenBSD, after a timeout, the kernel produces messages.
  4. A "good" result with garbage data, from a bad cable or bad memory in the electronics or the system.
I didn't see any of these. At least, no evidence has yet been presented that #4 has occurred.
Reply With Quote