On 4/19/2006 2:44 PM, Paul Dineen wrote:
I'm wondering if others have had difficulty with PDSE's which are included
in LNKLST at IPL time, specifically deletion of them?
There have been times where a LNKLST'd PDS needs expansion: the dataset can
be 'dequeued' via "SETPROG LNKLST,UNALLOCATE", stopping LLA and (in this
shop) MIM. Then a dataset can be reallocated larger under a new DSN,
populated, delete the old and rename the new as old. When done, restart
LNKLST allocation, LLA, MIM. Fairly mundane stuff.
I'd like to take this in a slightly different direction. First, I don't
understand why you don't simply use dynamic linklist to establish a new
entry in the linklist under a different name, and forget about the old
data set, until you do IPL.
But more than that, I'm curious about your comment about stopping MIM
and this being "mundane stuff".
I would have expected that by stopping MIM you would lose multi-system
serialization, and at that point you would need to ensure that no jobs
or users were active on the system. And at that point you might as well
IPL.
Are you working in a single-system environment as opposed to
multi-system? Or am I not understanding something else about your
environment? Or is this another scenario I should warn my colleagues
about that can cause corruption of the RACF database, so we can ask our
customers who use MIM about it on the rare occasion where they report
database corruption?
Walt Farrell, CISSP
z/OS Security Design, IBM
----------------------------------------------------------------------
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
