1 |
On Fri, 08 Jul 2011 19:40:58 -0400, Albert Hopkins wrote: |
2 |
|
3 |
> > Apart from the need to access legacy data, which Harry has resolved by |
4 |
> > reformatting, is there any benefit in using encfs rather than the |
5 |
> > in-kernel ecryptfs these days? |
6 |
> |
7 |
> Admittedly there isn't much difference, so if what you are using works |
8 |
> for you why not stick with it. |
9 |
|
10 |
Fair enough, except this thread is about encfs not working :( |
11 |
|
12 |
> I still prefer encfs, although I have |
13 |
> admittedly never tried ecryptfs, for the following reasons: |
14 |
> |
15 |
> * It's FUSE, completely userspace and requires no kernel support |
16 |
> (other than FUSE) and no special privileges to mount (other than |
17 |
> fusermount). |
18 |
|
19 |
On the other hand, it does seem quite a bit slower. Also, it means that |
20 |
your encrypted files much be user readable (IIRC) so someone could copy |
21 |
them for a more leisurely attempt at reading them. |
22 |
|
23 |
> * You can have multiple layers of encryption on on source |
24 |
> directory. E.g. two different passwords can give you two |
25 |
> different views of the filesystem. |
26 |
|
27 |
That's interesting, a bit of an edge case but interesting nonetheless. |
28 |
|
29 |
> * In the documentation at least, it says when you upgrade ecryptfs |
30 |
> you should first copy the files from the old ecryptfs to an |
31 |
> unencrypted filesystem, and then copy it to the new ecryptfs. |
32 |
> That seems like something some people won't want to do. |
33 |
|
34 |
That does seem a major drawback. I've not used ecryptfs that much, |
35 |
although I use to use encfs. I did try using it on a remote box on top of |
36 |
sshfs and the performance was appalling. |
37 |
|
38 |
|
39 |
-- |
40 |
Neil Bothwick |
41 |
|
42 |
"We are Microsoft of Borg. Prepare to...." |
43 |
The application "assimilation" has caused a General Protection Fault |
44 |
and must exit immediately. |