Re: ext3 status?

From: Riley Williams (rhw@MemAlpha.cx)
Date: Fri Mar 03 2000 - 04:44:12 EST


Hi Jeff, Stephen.

> I am assuming that the journal is being handled as a file
> (since it's represented as an inode), so after the convert,
> if I create the journal file empty, or not at all, EXT3
> will auto-create it the first time the volume is mounted?

If my memory's correct, ext3 will ignore a partition without a
journal and let it drop through to ext2. I'm not sure what it
does with an empty journal, but would suspect something similar.

I believe ext3 requires that the journal file occupies 12k (three
4k memory pages) of CONSECUTIVE disk space, and your conversion
would need to locate and allocate that for ext3 conversion to
work.

> This is what I am doing at present with NetWare2NTFS
> conversion (create it empty) and the first time Windows
> 2000 mounts the converted volume, the journal and the first
> 16 meta files are created when the volume is mounted.

Is that also true the first time the Linux NTFS system mounts the
said converted volume? I would suspect not...

> (OT) BTW - the NTFS driver (write) in Linux has data
> corruption on W2K. I have reviewed the code, and the
> on-disk structures are WRONG in several places, and several
> of the internal attribute records are making incorrect
> assumptions about some of the record fields.

> What's there will cause severe data-corruption if you are
> switching between the two platforms with a single partition
> image, particularly with the later W2K NTFS implementation
> (though NT 4.0 seemed to be OK -- some problems with the
> journal). The NTFS driver we tested was using the earlier
> FS implementation (n/2), not the newer W2K formats.

> The first time W2K mounts an NTFS volume with 4.0 format,
> it will auto-convert the NTFS volume to the W2K Volume
> Manager formats under Windows 2000, including any stripe
> sets, and potentially make incorrect decisions about
> attribute assignment if a journal is present and the volume
> has been booted under NT 4.0 after Linux has written to it.

> Thought you might like to know. We contracturally cannot
> help fix the NTFS driver in Linux, but we can point out
> potential problems and explain them to the best of our
> understanding.

Since you state that "the on-disk structures are wrong", can you
at least advise what the correct structures are, or is that
subject to an NDA of some sort? Your comments suggest that the
details available to whoever maintains the NTFS driver (not me)
are at best out of date, at worst simply plain wrong.

Best wishes from Riley.

 * Copyright (C) 1999, Memory Alpha Systems.
 * All rights and wrongs reserved.

+----------------------------------------------------------------------+
| There is something frustrating about the quality and speed of Linux |
| development, ie., the quality is too high and the speed is too high, |
| in other words, I can implement this XXXX feature, but I bet someone |
| else has already done so and is just about to release their patch. |
+----------------------------------------------------------------------+
 * http://www.memalpha.cx/Linux/Kernel/

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



This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:14 EST