On Wed, 15 Aug 2007 15:40:26 -0500, Paul Gilmartin <[EMAIL PROTECTED]> wrote:
>... >But this is for integrity in case: > > Address space A does a BLDL, FIND, or NOTE. > > Address space B replaces or deletes the member that > address space A found. > > Address space A then POINTs at the member. >... And I'm sure our integrity was well protected, but it did extend what was going to be a brief outage. My plan for implementing a new release of some brand-x product was 1. stop execution of the product on all LPARs 2. tell users to stay out of the product's ISPF interface. 3. copy the product's target linklib into the linklist library 4. do an LLA, UPDATE= on all LPARs 5. start things back up. Instead, there was a step for allocating a new PDS, adding a STEPLIB to a few procs, and some blind trust that the old ISPF interface was compatable with the new release. If I had been using a PDS rather than a PDSE (as had been the previous case, and was again after the next set of IPLs) the procedure would have contained 3a delete all members from the library 3b compress it 3c copy from the target library into the linklist library Is there less of an integrity exposure there than with the PDSE? I don't see it. Maybe if I had done an LLA UPDATE right after 3a, but I could have done that with the PDSE, too. (In fact I might have. This was 4 or 5 years ago.) Pat O'Keefe ---------------------------------------------------------------------- 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
