On Sat, 5 Jan 2008 17:53:13 -0600, Patrick O'Keefe wrote: > > If you don't mind having incredibly convoluted RESTORE >processes, don't ACCEPT anything except the FUNCTION. There is an >obvious trade off between ACCEPT and RESTORE. > >Perhaps it would be a good enhancement to SMP (SMP/E2?) to provide >generations of levelsets. You're free to submit a request. > Tedious, but not convoluted. At each intended levelset point, LIST all the sysmods in the target zone. Then, based on the list, RESTORE everything except the FUNCTIONs (because PTFs with prerequisites can't be restored alone), and re-APPLY everything in the list.
Of course, some PTFs can't ever be RESTORED. ... >You are right, though. Using SMP you can go back only to the >previously ACCEPTed maintenance level. You can restore libraries >back to any arbitrary backup point. > Thereby leaving the libraries out of sync with the data base? Gasp! You're making me believe that SMP/E should keep checksums of all SMPPTS, TARGET, and DLIB members, and fail if they can't be verified, kind of like the recent thread about someone who tried to modify a program object using an unsupported method. >You can obviously do the same with backups of SMP datasets, target >libraries, and DLIBS (but you probably don't want to do that if you >have more than one product in the CSI). -- gil ---------------------------------------------------------------------- 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
