After reviewing the DoD specifications for destruction by overwriting, I would say that your method does not meet them. Specifications are available here: http://www.tricare.mil/tmis_new/ia/02%20-%20Sanitization.pdf , section 3.1.2. They specify that you must overwrite with a pattern, then the complement of the pattern, then with random data. They further specify that you must overwrite the entire disk, "independent of any BIOS or firmware capacity that the system may have". Among other things.
My question would be, do you really need to meet DoD specifications? If so, you'll probably need something like FDRERASE, which is certified to meet those specifications. > -------- Original Message -------- > Subject: ICKDSF Release 16 > From: Cliff Brenner <[EMAIL PROTECTED]> > Date: Wed, January 31, 2007 4:03 pm > To: IBMVM@LISTSERV.UARK.EDU > > Hi Folks, > > We are in the process of 'decommissioning' our mainframe platform > (MP3000 runing v/VM 3.1). We formatted all our internal and external > DASD (3380s and 3390s) using ICKDSF R16 with the following command: > > CPVOLUME FORMAT UNIT(nnn) NOVERIFY VOLID(Lnnn) - > where nnn is the real pack address > > We formatted most of the packs on VM, then shut down the system and > formatted the CP-owned packs using ICKDSF standalone. > > Now that the work is done, we are getting questions as to whether > ICKDSF formatting conforms to certain Department of Defense standards > which recommends multiple formats to guarantee all data has been > removed. > > The ICKDSF manual reads that CPVOLUME FORMAT writes CP information > to cylinder 0 and then lays out 4K pages on the remaining cylinders. > Does anyone know whether this means 4K pages comprising of binary > zeroes? Since our DASD are CKD, what happens to any tracks (if any) > that don't fall into 4K boundaries? Is formatting a pack once good > enough to insure all data is irrecoverable? Any assistance with these > questions would be greatly appreciated. Thank you. > > Cliff Brenner > Pace University Computer Systems Dept.
