In a recent note, "(IBM Mainframe Discussion List)" said: > Date: Mon, 22 Jan 2007 10:52:29 EST > > In a message dated 1/22/2007 8:16:41 A.M. Central Standard Time, > [EMAIL PROTECTED] writes: > >think IBM's answer to the security issue is an erase feature. > There is no other way to enforce the requirement that user A's data cannot > be read by user B after user A has released ownership of the tracks and user B > subsequently is allowed to allocate the same tracks. Just don't start > Sure there is. Prohibit reading beyond DS1LSTAR, and enforce management of DS1LSTAR so it is never allowed to point past uninitialized space.
> erasing all tracks in all data sets willy-nilly, or you may have DASD > performance > problems like you wouldn't believe. Be VERY selective about what you erase. > OTOH, writing an EOF at the beginning of every newly allocated extent _regardless_of_DSORG_ would add little performance burden, and make behavior more predictable, but it wouldn't address the data security issue. Isn't there an option to erase on freeing an extent? The MVS progeny are extraordinary among existing operating systems in failing to keep track of where files end. -- gil -- StorageTek INFORMATION made POWERFUL ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
