1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
|
*** 1 ***
The edit/write inode stuff is still very much ALPHA. So far I've
tested ext2, minix, and xiafs on a 172k partition and they worked
under minimal testing. I copied the partition to a file, modified one
block pointer, wrote it, then did a "cmp -l" on the partition and the
copy and it reported only one block different (with the correct values
even). Not bad, but I would appreciate it if someone could verify
something similar on a larger scale, particularly the ext2fs.
*** 2 ***
There is a bug/misfeature in the Linux 2.0 kernels that will make
undelete fairly hard. When files are deleted, all indirect blocks are
zeroed. I wrote a small kernel patch and used it successfully for
most of last year, but according to Ted, it probably won't work.
Looking it over, I think he's right, so I've stopped using and
distributing it. Most of the kernel developers are working on Linux
2.1, but if anyone has a fix for 2.0 or wants to work on one, email me
(sdh@po.cwru.edu).
|
