Re: [PATCH RFC] xfs: remap block layer ENODATA read errors to EIO

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 





On 22/8/25 06:06, Eric Sandeen wrote:
On 8/21/25 7:52 AM, Carlos Maiolino wrote:
On Thu, Aug 21, 2025 at 07:16:49PM +1000, Donald Douwsma wrote:


On 19/8/25 06:45, Darrick J. Wong wrote:
On Mon, Aug 18, 2025 at 03:22:02PM -0500, Eric Sandeen wrote:
We had a report that a failing scsi disk was oopsing XFS when an xattr
read encountered a media error. This is because the media error returned
-ENODATA, which we map in xattr code to -ENOATTR and treat specially.

scsi_debug can be configured to return a MEDIUM error which if I follow
the discussion properly, would result in the block layer converting it
to ENODATA.

Yep that does work though very old scsi_debug modules can't control
which sector returns ENODATA. But for semi-modern scsi_debug it should
work quite well.

-Eric


Thanks Carlos, Eric, that works well.

I should able to make it work without knowing the sector if I stat the
inode before enabling the error if we need support for older kernels.

I'll file off the rough edges and post it for review.

Don






[Index of Archives]     [XFS Filesystem Development (older mail)]     [Linux Filesystem Development]     [Linux Audio Users]     [Yosemite Trails]     [Linux Kernel]     [Linux RAID]     [Linux SCSI]


  Powered by Linux