Comment 10 for bug 242448
Revision history for this message
| Serge Hallyn (serge-hallyn) wrote Re: [Bug 242448] Re: Hardy ecryptfs can cause data loss | #10 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
0e1f616
(Get the code!)
On Tue, Aug 12, 2008 at 8:42 AM, Adam Wood <email address hidden> wrote:
> I've just gone back to test it again and have discovered there may be a
> time element involved.
>
> After my earlier test I accidentally forgot to unmount /tmp/dir2
> I've just unmounted and re-mounted it and there was file corruption.
>
> I've just repeated the test using 2 new directories, /tmp/dir3 and
> /tmp/dir4. Unmounting and remounting immediately shows that they are
> decrypting fine.
Just to be clear, since you said 'decrypting fine': if you read the
files, they will decrypt fine. It is only if you append to the file
immediately after mount, and the file is more than one block in size
and not an exact multiple of blocksize (iirc).
If you did mean using append, then that is interesting, bc I've never
had it fail to corrupt a file :) (or is that :( ).
-serge