Hi,
Yes, the Seagate FreeAgent USB drives issue is well-known since April but
this one is quite new and more and more people are reporting.
It seems to be either kernel device driver bug or the market was flooded
with very low quality disks, as sometimes this already happened in the past
too. E.g. here are the same problem with other LaCie disks:
http://ubuntuforums.org/showthread.php?t=574942
[ 128.312000] sd 2:0:0:0: [sdb] Result: hostbyte=DID_ERROR
driverbyte=DRIVER_OK,SUGGEST_RETRY
[ 128.312000] end_request: I/O error, dev sdb, sector 490234687
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234624
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234625
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234626
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234627
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234628
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234629
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234630
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234631
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234632
[ 128.312000] Buffer I/O error on device sdb1, logical block 490234633
[ 791.420000] sd 2:0:0:0: [sdb] Result: hostbyte=DID_ERROR
driverbyte=DRIVER_OK,SUGGEST_RETRY
[ 791.420000] end_request: I/O error, dev sdb, sector 490234687
The errors happen __right__ after mount, so the disk doesn't have
even one second idle time to trigger a power down.
The problems are always with the $Bitmap file which is obviously
intensively used. I'm getting afraid that the media on the disk
surface may wear out too early and causing these problems.
Szaka
On Mon, 29 Oct 2007, Bernhard Kaindl wrote:
> On Mon, 29 Oct 2007, Szabolcs Szakacsits wrote:
> >
> > Hola Rico,
> >
> > On Mon, 29 Oct 2007, Rico Lopez Rafael wrote:
> >
> > > ntfs_attr_pread: ntfs_pread failed: Input/output error
> > > Failed to read NTFS $Bitmap: Input/output
>
> I got the similar messages when using Seagate FreeAgent USB drives which
> spin down their disk after 15 minutes of inactivity by default.
>
> In /var/log/messages, above these these ntfs errors kernel: sd 5:0:0:0: [sdc]
> Device not ready: <6>: Sense Key : Not Ready [current]
> kernel: : Add. Sense: Logical unit not ready, initializing command required
> kernel: end_request: I/O error, dev sdc, sector 6291527
> kernel: Buffer I/O error on device sdc1, logical block 786433
> ntfs-3g: ntfs_attr_pread: ntfs_pread failed: Input/output error
> ntfs-3g: Error reading $Mft record(s): Input/output error
> ntfs-3g: ntfs_mft_record_read failed: Input/output error
>
> Since an Input/output error (I/O error) is normally an indication of a
> problem actually doing the read/write operation on the device, it sounds
> like you should also check the kernel logs for any USB or SCSI errors
> like those I got.
>
> While I am not aware of any linux kernel issues with LACIE Big Disk USB 500GB
> drives, but I thought I'd tell you that you should look there as well. The
> command
> 'dmesg' shows you the latest messages which are logged by the kernel.
>
> The problem with the Seagate FreeAgent USB drives will hopefully be solved
> for all future with the next upstream kernel release (2.6.24), at least a
> fix which enables USB disk startup in case it is not ready was in GregKH's
> patch list for the new kernel the last time I saw it.
>
> If you see kernel errors, you can also try a newer kernel (at your own risk),
> and the fix for the Seagate FreeAgent USB is to automatically enable extended
> SCSI error recovery by setting "allow_restart" (can also be set thru sysfs)
> for all USB disks. It causes that USB disks which are not ready for I/O can be
> tanspartently started up when they are accessed thru the kernel's SCSI layer.
>
> Bernhard
>
> >
> > Run 'chkdsk /f DRIVE:' on Windows and retry (Windows bug).
> >
> > If mount fails again then run 'chkdsk /f /r DRIVE:' on Windows and retry
> > (hardware fault).
> >
> > If it still fails then please send me the NTFS metadata:
> >
> > http://www.linux-ntfs.org/doku.php?id=ntfsclone#store_only_ntfs_metadata
> >
> > Thanks,
> > Szaka
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
ntfs-3g-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel
