On Fri, May 31, 2013 at 8:59 PM, Paul Gilmartin <[email protected]> wrote: > On Fri, 31 May 2013 21:24:06 -0400, John Gilmore wrote: >> >>The production-library member must be identical to the acceptance-test >>library member, and the only way to ensure that this is the case is to >>copy the [successful-outcome] acceptance-test member into the >>production library. [The binder is, of course, very much better at >>such member-copying operations than any of the alternatives to it; and >>it should always be used.] >> > One tool to verify identity is the checksum. Will the binder so "copy" > members that checksums can be recreated? If not, I consider the > operation to have some of the character of a recompilation reather > than a "copy". And is there any way to verify that the metadata > resulting from a "copy" by the binder are identical? > > In the open world, suppliers regularly (though not always) supply > checksums that recipients can verify. SMP/E network installation > verifies checksums over part of the process (though far from > end-to-end). > > -- gil > On the Mainframe, most programs have the date and time of the compile stored in the eyecatcher near the name. Sometimes for a release they change these to be the same. They often also add a PTF name into the eyecatcher. -- Mike A Schwab, Springfield IL USA Where do Forest Rangers go to get away from it all?
---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
