On Fri, 23 Apr 2010 17:01:29 -0400, Robert A. Rosenberg wrote: >At 10:54 -0500 on 04/23/2010, Tom Marchant wrote about Re: ACCEPT and >RELFILES deleted: > >>I have always considered the possibility of cloning an additional DLIB zone >>for those cases where I needed to restore some maintenance but couldn't >>because of prereqs. > >This points out what is IMO a MAJOR design flaw in the way RESTORE >works. If I want to restore a SYSMOD, it should be able to restore >the elements in the SYSMOD by selecting the copies that were replaced >by the SYSMOD without needing to use copies of other elements in the >SYSMODs that have the needed elements. > I might say "deficiency" rather than "flaw".
>IOW: If SYSMOD1 contains E1 and E2 (and PREs SYSMOD2 which contains >E1 and E3 and SYSMOD3 which contains E2 and E4) then to RESTORE >SYSMOD1 just use E1 from SYSMOD2 and E2 from SYSMOD3 WITHOUT needing >to RESTORE SYSMOD2 and SYSMOD3 (and their PRE/SUP chain until E1, E2, >E3, E4 [and all the other elements that are conatined in the chain]) >have been RESTORED followed by an APPLY of all the SYSMODs (except >for SYSMOD1) which were just uselessly RESTORED due to the poor >design of RESTORE. IMO, a RESTORE should be done by doing the >equivelent of an APPLY of ONLY the elements that are being backed off >WITHOUT needing to touch any element that is not contained in the >SYSMOD being RESTORED. > Well, if you ACCEPT SYSMOD2 and SYSMOD3, you can do that. But that locks you out from ever RESTOREing SYSMOD2 or SYSMOD3 subsequently. What you want is some sort of UNDO, which SMP/E doesn't provide, even assuming the best case, in which SYSMOD2 and SYSMOD3 have not been PURGEd, so E1 and E2 are available somewhere in the GLOBAL zone (SMPPTS or SMPTLIB). SMP/E just won't deign to get them. The SMP/E partisans will chime in at this point and say that sort of operation was never a design objective of SMP/E. But why not? VMFMERGE allows that; it has no analogue of ACCEPT; it just fetches E1 and E2 in proper versions from the DELTA disk, its analogue of the SMPPTS. I consider ACCEPT a hindrance rather than a useful function. -- 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
