The DoD 5220.22-M document, which is often allegedly "quoted" for describing the 3-pass wiping on hard-disks, distinguishes the two following situations (see section 8-301):
- Clearing. Clearing is the process of eradicating the data on media before reusing the media in an environment that provides
an acceptable level of protection for the data that was on the media
before clearing. All internal memory, buffer, or other reusable memory
shall be cleared to effectively deny access to previously stored
information.
- Sanitization. Sanitization is the process of removing the data from media before reusing the media in an environment
that does not provide an acceptable level of protection for the
data that was in the media before sanitizing. IS resources shall
be sanitized before they are released from classified
information controls or released for use at a lower
classification level.
Sanitization will therefore very often lead to the media destruction. However, here in this post we are merely focusing on clearing.
I will share below my few findings on the subject.
Data deletion is handled by the disk writer, not the software
Disk erasing is somewhat similar to HDD's SecureErase feature mentioned in the question. Indeed, when asking a software to erase a disk, all it will do is send an ATAPI BLANK message to the disk writer's firmware, and it will be up to the firmware to handle actual data erasing and send back the operation result to the software.
However, unlike the SDD, I did not encountered any information leading to think that some writer may be more or less efficient than other to erase a rewriteable disk content.
There is two level of erasing
Some writers actually allow even more fine-grained data erasing, like per-track erasing, but as per our concern it is very interesting to note that they propose at least two erasing level:
- A quick erase, used by default when erasing a disk, will only delete the disk's Table Of Content (TOC) and a few other related information, but the data themselves will remain on the disk. It seems very similar to a quick HDD formatting. I also strongly suspect that a lot of end-user software may only propose this mode.
- A full erase will take a long longer since the software will instruct the firmware to go through all the disk surface to erase all data from it to turn back the disk into pristine state.
Full disk erase are efficient
My few searches did not point out any weakness to a rewriteable disk full erase. On the contrary, all forensics vendor I went through seem to limit efficiency of their products (whether it is software or hardware) to quick-erased disks.
While from a pure theoretical point-of-view one may still imagine some tiny weakness exploitable by a large and well-founded organization, as a reminder we are here in the scope of data clearance and not sanitization. For clearance goal, a full disk erase is therefore fine, the trick being to not mistakenly use a quick erase where, on the contrary, the data could be retrieved relatively easily.
How to proceed
On Unix platforms, a full disk erase can be achieved using the cdrecord tool. It can be used the following way:
cdrecord blank=all dev=/dev/scd0
blank=all: to be opposed to blank=fast, this the actual argument which will initiate the full disk erase process,dev=/dev/scd0: to indicate the path to your disk writer device,-force: this usefull parameter can be added to allow cdrecord to use "some tricks" which may allow to erase a disk which would not be erasable otherwise (damaged disks for instance). cdrecord manpage.
