Paul Gilmartin wrote:
On Sat, 12 Jan 2008 17:55:27 -0500, Robert A. Rosenberg wrote:
My biggest gripe with the design of SMP/E is the use of a RESTORE
design that will back-out a PTF (or set of PTFs) to get back to the
state you were in (or would have been in) before you APPLY'ed the
PTF(s). Right now you must RESTORE PTFs that you will just turn
around a reAPPLY just to RESTORE a PTF that PREs or SUPs the PTFs. A
better design is to see what SYSMOD owns each element that is being
RESTOREd and just do an automatic APPLY of only that element instead
of removing elements that are not contained in the PTF being RESTOREd.
I agree wholeheartedly. RESTORE should be "Undo", as we know it
on our desktop systems. I grant that it's logically impossible
to RESTORE to a level of an element that's been PURGEd from the
SMPPTS; otherwise, if the content is in the SMPPTS (and/or TLIBs)
it should be permitted to RESTORE it. I suspect the CSI simply
fails to retain information needed for this task.
SMP/E is too efficient! It's design point is, when performing a "mass"
apply of hundreds (or even thousands) of PTFs, that it installs only the
highest level of each element being replaced. Intermediate levels are
neither created nor journaled. There is no version to "undo" to!
Backing off any one PTF requires a RESTORE (from DLIB) of all elements,
from all interdependent PTFs, followed by re-APPLY specifying EXCLUDE
for the "problem" PTF. Granted, it would be nice if more of this
processing could be performed automatically.
--
Edward E Jaffe
Phoenix Software International, Inc
5200 W Century Blvd, Suite 800
Los Angeles, CA 90045
310-338-0400 x318
[EMAIL PROTECTED]
http://www.phoenixsoftware.com/
----------------------------------------------------------------------
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
