Forum Moderators: bakedjake
rm * -r -f
your home, log, and etc directories first and then cd / and do the same from there. The OS would probably get about halfway before realising that it's eating itself.
Same applies to any mkfs command I guess.
CompWorld:
I've never done this before, but I would unmount the partition and then write random data to it. Something like dd if=/dev/urandom of=/dev/sda1 would probobly do it. I don't know if you need to add some more argument to dd (to tell dd how big the partition is ie.).
The problem is that you cannot unmount root /usr and /var because the system is using them. I would delete all secret files from those partitions and then fill the partitions with random files. Then all the secret data should be overwritten I guess.
I've never done anything like this before, but since you didn't get any other response I wrote this post.
If this is a hosting company we're talking about, then this may not be possible, if you're not root (and you should check whether you're allowed to do this).
Basically you can run "dd if=/dev/zero of=/dev/hda0" (replace hda0 with whatever your disk's ID is). DO NOT close the connection... You may not get it back... SEe if there's a keepalive option or anything of the like.
I would use the wipe utility or similar just on /home and /var/log, then once done you can set off a final rm -rf for the rest, and if something is left behind once the connection breaks it would not be a big problem.
How are you going to destroy your host's backup tapes?
With a team of men in black ski masks? LOL
That's a very good point.
I should not though, that as well as that fact, there's also the fact that this box has been insecure from the start, so how could you be certain of anything? I don't think that doing a wipe of any kind will help you in terms of securing any previously written data.
If you truly want to be paranoid in the future, run everything from your own box, with encrypted drives, etc...