[lug] Read-only filesystem

[Gary Hodges]

Hugh Brown wrote:
> 
> On Wed, 3 Jan 2007, Gary Hodges wrote:
> 
>> Jeff Schroeder wrote:
>>> Gary wrote:
>>>
>>>> When copying data to one of the disks I
>>>> will eventually get an error that says in part "cannot create regular
>>>> file" "read-only file system."
>>> If there's a hardware failure in the hard drive, the kernel will
>>> sometimes instruct 'mount' to change the filesystem to read-only in
>>> order to prevent additional damage.  This might be what's happening;
>>> check your kernel logs or even 'dmesg' to see if you're getting reports
>>> of bad sectors or something along those lines.
>> Thanks for the reply.  I see these messages with dmesg:
>>
>> [ 2890.878069] sd 6:0:2:0: SCSI error: return code = 0x8000002
>> [ 2890.878073] sdc: Current: sense key: Aborted Command
>> [ 2890.878075]     Additional sense: Scsi parity error
>> [ 2890.878081] end_request: I/O error, dev sdc, sector 156253039
>> [ 2891.291088] REISERFS: abort (device sdc1): Journal write error in
>> flush_commit_list
>> [ 2891.291094] REISERFS: Aborting journal for filesystem on sdc1
>>
> 
> Yep, you've got a bad drive or a bad controller.  Get the data off the
> disk if possible.

Thank you for the reply.  I copied everything to another disk and 
thought I would attempt some disk tests.  The Fujitsu utility only runs 
on Windows and I don't have a SCSI Windows to put the disk in so I 
started looking at smartctl.  I just finished running a long test <sudo 
smartctl -c -t long /dev/sdc> and here are the results:

=======================
~>sudo smartctl -l error /dev/sdc
smartctl version 5.36 [x86_64-unknown-linux-gnu] Copyright (C) 2002-6 
Bruce Allen
Home page is http://smartmontools.sourceforge.net/


Error counter log:
            Errors Corrected by           Total   Correction 
Gigabytes    Total
                ECC          rereads/    errors   algorithm 
processed    uncorrected
            fast | delayed   rewrites  corrected  invocations   [10^9 
bytes]  errors
read:          0        1         0         0          0       6782.744 
           0
write:         0        0         0         0          0        179.515 
           0

Non-medium error count:     1164
=======================

If I did this correctly, it looks like the drive passed.  Are you sure 
the dmesg output above indicates a failing drive?  I believe the drive 
is under warranty and I can do without it for a week or so I'll probably 
  go ahead and send it in.