[smartmontools-support] Extended test doesn't catch CurrentPendingSector and OfflineUncorrectableSector errors
Christian Franke
Christian.Franke at t-online.de
Mon Jul 12 19:40:10 CEST 2021
Dipanjan Das wrote:
>
> Hi Christian,
>
> Thanks for your detailed response.
>
> On Mon, 12 Jul 2021 at 04:14, Christian Franke
> <Christian.Franke at t-online.de <mailto:Christian.Franke at t-online.de>>
> wrote:
>
>
> This log matches the ddrescue result. Read of (at least) LBA
> 3884004178
> failed occasionally and worked again in later tests. Unfortunately
> the
> disk firmware did not redirect this weak sector.
>
>
> (197) Current_Pending_Sector is set to 1 for the last few weeks. Maybe
> that sector refers to this one where the read fails intermittently?
Possibly, see below.
> I have run several extended checks since this error appeared,
> hoping that it would get remapped, and (5) Reallocated_Sector_Ct will
> bump up. Unfortunately, that didn't happen. Not sure if it's because
> of the intermittent nature of the failure.
Overwriting the sector with same data should reallocate it.
Or: Check whether a file is affected (e.g. with sleuthkit). If any, copy
the file to somewhere else and overwrite the original in-place (e.g.
with shred).
>
> In the SMART attributes section, (198) Offline_Uncorrectable is set to
> 6 for the last few weeks. Also, I am curious what's that related to.
Some drives do not clear this attribute on reallocation.
Some drives do not bump Reallocated_Sector_Ct on reallocation.
SMART attributes are not part of any ATA standard, so vendors could use
any interpretation.
> ...
>
> Attached. Didn't pass '-q noserial' intentionally if that helps you in
> any way.
> 9 Power_On_Hours -O--CK 026 026 000 - 54399
> ...
> 12 Power_Cycle_Count -O--CK 100 100 000 - 49
Operating 24/7 for 6+ years ?
I would suggest to replace this drive :-)
> After command completion occurred, registers were:
> ER -- ST COUNT LBA_48 LH LM LL DV DC
> -- -- -- == -- == == == -- -- -- -- --
> 40 -- 51 00 80 00 e7 33 00 81 52 40 00 Error: UNC at LBA =
> 0xe733008152 = 992993116498
This firmware uses the wrong byte order in this error log. Please retry with
smartctl -l xerror -F xerrorlba ...
This may result in "LBA = 0xe7813352 = 3884004178" which then matches
the LBA from the self-test log. This is possibly the Current_Pending_Sector.
Regards,
Christian
More information about the Smartmontools-support
mailing list