New variant of ext2 patch is available. Patch is against -test9
and -test10-pre1 and it lives on ftp.math.psu.edu/pub/viro/ext2-patch-9e.gz
Interface between namei.c and dir.c remains the same, dir.c got
serious cleanup compared to the previous version. Handling of records with
wro
On Thu, 28 Sep 2000, Andreas Dilger wrote:
> This would also explain why the direntry offset is pointing past the
> end of the first dir block, because the next block was added at the time
> the new file was created, but then later lost.
Nope. Try the following:
@@ -378,12 +378,13 @@
Marco writes:
> On Thu, Sep 28, 2000 at 11:51:02AM -0400, Alexander Viro wrote:
> > On Thu, 28 Sep 2000, Marko Kreen wrote:
> > > On Tue, Sep 26, 2000 at 05:29:27PM -0400, Alexander Viro wrote:
> > >
> > > EXT2-fs error (device sd(8,6)): ext2_add_entry: bad entry in directory
> > > #28699: rec_le
On Thu, 28 Sep 2000, Marko Kreen wrote:
> root@snoopy:~# debugfs
> debugfs 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
> debugfs: open /dev/sda6
> debugfs: stat lib/postgres/data/base
> Inode: 14479 Type: directoryMode: 0700 Flags: 0x0 Generation: 883245
> User:31 Group:
On Thu, Sep 28, 2000 at 11:51:02AM -0400, Alexander Viro wrote:
> On Thu, 28 Sep 2000, Marko Kreen wrote:
> > On Tue, Sep 26, 2000 at 05:29:27PM -0400, Alexander Viro wrote:
> >
> > EXT2-fs error (device sd(8,6)): ext2_add_entry: bad entry in directory
> > #28699: rec_len is smaller than minimal
On Thu, 28 Sep 2000, Marko Kreen wrote:
> On Tue, Sep 26, 2000 at 05:29:27PM -0400, Alexander Viro wrote:
> > Folks, give it a try - just keep decent backups. Similar code will
>
> It started to give me following errors:
>
> EXT2-fs error (device sd(8,6)): ext2_add_entry: bad entry in dir
On Thu, 28 Sep 2000, Marko Kreen wrote:
> On Tue, Sep 26, 2000 at 05:29:27PM -0400, Alexander Viro wrote:
> > Folks, give it a try - just keep decent backups. Similar code will
>
> It started to give me following errors:
>
> EXT2-fs error (device sd(8,6)): ext2_add_entry: bad entry in dir
On Thu, Sep 28, 2000 at 01:56:48PM +0200, Marko Kreen wrote:
> Hmm, I am starting to think that maybe I _should_ make backups...
>
The feeling is even stronger now...
I dropped into single-user mode and did a fsck to /var.
fsck gave me a lot of errors in style
'unattached inode XXX ' 'inode
On Tue, Sep 26, 2000 at 05:29:27PM -0400, Alexander Viro wrote:
> Folks, give it a try - just keep decent backups. Similar code will
It started to give me following errors:
EXT2-fs error (device sd(8,6)): ext2_add_entry: bad entry in directory
#28699: rec_len is smaller than minimal - offs
On Wed, 27 Sep 2000, Ingo Molnar wrote:
>
> On Wed, 27 Sep 2000, Marko Kreen wrote:
>
> > > Why?
> > >
> > > > + } else if (de->name[2])
> > >
> > Sorry, I had a hard day and I should have gone to sleep already...
>
> hey, you made Alexander notice an endianne
On Wed, 27 Sep 2000, Marko Kreen wrote:
> > Why?
> >
> > > + } else if (de->name[2])
> >
> Sorry, I had a hard day and I should have gone to sleep already...
hey, you made Alexander notice an endianness bug so it was ok :-)
Ingo
-
To unsubscribe from th
On Tue, 26 Sep 2000, Alexander Viro wrote:
> On Tue, 26 Sep 2000, Andreas Dilger wrote:
>
> > Al Viro writes:
> > > Folks, give it a try - just keep decent backups. Similar code will
> > > have to go into UFS in 2.4 and that (ext2) variant may be of interest for
> > > 2.4./2.5. timeframe.
>
On Tue, 26 Sep 2000, Andreas Dilger wrote:
> Al Viro writes:
> > Folks, give it a try - just keep decent backups. Similar code will
> > have to go into UFS in 2.4 and that (ext2) variant may be of interest for
> > 2.4./2.5. timeframe.
>
> Haven't tested it yet, but just reading over the p
Al Viro writes:
> Folks, give it a try - just keep decent backups. Similar code will
> have to go into UFS in 2.4 and that (ext2) variant may be of interest for
> 2.4./2.5. timeframe.
Haven't tested it yet, but just reading over the patch - in ext2_lookup():
if (dentry->d_name.len
On Tue, Sep 26, 2000 at 06:31:04PM -0400, Alexander Viro wrote:
> On Wed, 27 Sep 2000, Marko Kreen wrote:
> > There is something fishy in ext2_empty_dir:
>
> Why?
>
> > + } else if (de->name[2])
>
Sorry, I had a hard day and I should have gone to sleep already...
I
On Wed, 27 Sep 2000, Marko Kreen wrote:
> On Tue, Sep 26, 2000 at 05:29:27PM -0400, Alexander Viro wrote:
> > Comments and help in testing are more than welcome.
>
> There is something fishy in ext2_empty_dir:
Why?
> + /* check for . and .. */
> +
On Tue, Sep 26, 2000 at 05:29:27PM -0400, Alexander Viro wrote:
> Comments and help in testing are more than welcome.
There is something fishy in ext2_empty_dir:
+ /* check for . and .. */
+ if (de->name[0] != '.')
+
Duh. After fixing two idiotic bugs in ext2_readdir() it seems to
be really working (survives assorted builds, does the right thing on
find-based scripts and obvious local tests, yodda, yodda). It certainly
needs more testing, but I would call it (early) beta.
Folks, give it a try
OK, it seems to be more or less working. Patch is available on
ftp://ftp.math.psu.edu/pub/viro/ext2-patch-6.gz. It's against -test9-pre7.
Help in testing is welcome, just keep in mind that it's ext2 we are
talking about. IOW, proceed with care and don't let it loose on the data
you can't e
19 matches
Mail list logo
