On Wed, 3 Oct 2007, Timothy Shimmin wrote:
Hi Justin,
Justin Piszcz wrote:
On Tue, 2 Oct 2007, Lachlan McIlroy wrote:
Yeah that about sums it up. In an attempt to prevent log replay of inodes
in cases when we shouldn't replay we also prevented log replay of inodes
in
cases when we should replay. We end up with directories that refer to
inodes
that were not replayed and we read existing data off disk. That existing
data is usually previous instances of inodes. We had cases of regular
files
turning into directories and inode version mismatches.
Lachlan
In 2.6.23-rc8?
The bad change went into rc7 and was still there in rc8.
It was backed out for rc9.
--Tim
If one with was running 2.6.23-rc8 with XFS, does that mean we should run
xfs_repair on our filesystems after upgrading to -rc9?
Justin.
|