Re: Deleted files recovery
david parsons (o.r.c@p.e.l.l.p.o.r.t.l.a.n.d.o.r.u.s)
14 Sep 1998 11:22:32 -0700
In article <linux.kernel.19980912162855.13343@bug.ucw.cz>,
Pavel Machek <pavel@bug.ucw.cz> wrote:
>Hi!
>
>> My informal undelete solution is to turn the power off (or hit the
>> reset or suspend switch) as soon as I realize I made a stupid
>> mistake. This works for me about 70% of the time (5 seconds is a
>> _really_ long time after you've wipe out a module you've been working
>> on for 18 hours.) though it didn't help the time I miscoded something
>> in McAfee's Virus Scan and ended up eating /home/orc on my development
>> laptop [thank goodness for the ext2fs editor.]
>
>..this is exactly why I set my update to 30 seconds - so that i have
>more time to turn it off.
>
>
>> This thing needs to have a tool that can manipulate whiteout entries,
>> and a daemon that can reap whiteout entries when the disk gets full,
>> but those are pretty trivial compared to keeping and having to
>> rewrite applications to understand .deleted directories (when talking
>> about tweaking filesystem intrinsics, userspace solutions suck on
>> monolithic kernels.)
>
>Actually, making use of .deleted directories is pretty trivial, you
>have just to play with LD_PRELOAD a bit.
I carry around statically linked objects (both a.out and elf, for
distribution), and dynamically linked objects (both a.out and elf,
for personal use.)[1] Carrying around a pair of preload libraries
(three, if I was foolish enough to use libc 6) seems like a recipe
for inconsistant behavior.
____
david parsons \bi/ [1: `Don't use a.out' is the wrong answer.]
\/
-
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/faq.html