Re: PROBLEM: Reiserfs stall 2.6.10-bk7 up through 2.6.12-rc3

[George Ronkin]

On Fri, 2005-05-06 at 18:13 -0700, George Ronkin wrote:
> ... one of my machines now stalls
> attempting to access the reiserfs file system used as that server's
> spool...  This problem did not occur before 2.6.10-bk7.  That and all subsequent
> kernels I've tried (Debian 2.6.11, stock 2.6.12-rc3, and 2.6.12-rc3-mm3)
> cause the problem, consistently and repeatably.
	As does 2.6.12-rc4.  The unusual 1K block size I used for this reiserfs
seems to have exposed the problem; all my other reiserfs use 4K and have
no problem.  I copied its contents to a new reiserfs I created with 4K
blocks and the later kernels work fine with the copy.  I also tried
turning off CONFIG_QUOTA on 2.6.12-rc4 - that worked with the 1K block
reiserfs as well.  I'm keeping the 4K block copy, since that works with
QUOTA, and turning QUOTA off affects non-reiserfs fs as well.
	Note also:

- The 1K block fs caused the problem even though no quota mount options
were set.

- All my reiserfs are devmapped, so I don't know if the problem occurs
without it on a physical partition, or whether the symptoms would
differ.
	HTH,
-- 
George Ronkin <gronkin@xxxxxxxxxxxx>

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/