Tom Marchant wrote:
The oldest GDS will roll off regardless of SCRATCH. If neither
SCRATCH nor EMPTY is specified, the oldest GDS will be in rolled-out
status and will no longer be part of the GDG. In the pre-SMS days
it would have been uncataloged but not scratched. That said, I
agree with David that *the GDG should be defined with SCRATCH.
*
This seems to be one of those weird SMS scenarios that crop up from time
to time; either you spend "days" trying to work out exactly what caused
the problem or you correct the problem by normal means and assume that
this was an unforeseen situation caused when X and Y happened at the
same time that SMS or HSM was doing Z.
NOSCRATCH might be useful if you easily need to process a GDG after the
limit on the rolled-in GDSes has been exceeded following a failure where
you are processing data that arrives at +1.
Also there is a limit in the number of GDS names that can exist:
http://www-1.ibm.com/support/docview.wss?uid=isg1II07276
and peculiar effects can be observed if you have "holes" in your GDS
numbers. This is how we discovered that our SMS settings for GDGs and/or
GDSes were wrong.
----------------------------------------------------------------------
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
