Charles Mills wrote:
99% sure the answer is yes. I know it is possible to open a new dataset and
ATTEMPT to read the data already on the tracks. My tests have tended to
"fail" (so to speak) for the reason that I am not "trying" to read someone
else's data, and the tests tend to fail on bad block size (if reading FB) or
invalid RCW format (if reading VB). I suspect an attempt reading RECFM=U
would succeed.
I think IBM's answer to the security issue is an erase feature.
Charles
</snip>
I was successfull in reading residual data on disk may years ago and
consequently wrote an allocation exit that wrote an EOF record for all
sequential allocations. The blocksize just had to be larger than that of
the residual data. Since then, from my dodgy memory, writes an EOF
record for all new sequential allocations, PDS did not need this exit as
it writes an empty directory. It may still be possible to read residual
data for non SMS sequential allocations, but I have not tried it recently.
Regards,
Paul Gillis
----------------------------------------------------------------------
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
