CC's response was as usual completely correct. By the way, I believe that it is stated somewhere not to have this situation.. And that means that how the system behaves is undefined. You are welcome to hope that the actual behavior matches what you want.
The only case of interest is when something actually requests load of "WOMBAT". Otherwise this is no different than a scenario where no alias at all was involved. So consider that same setup where both FRED and JOE are loaded, and then someone loads WOMBAT. Though not stated in the post, I'll assume that FRED is in an earlier data set in the concatenation than JOE since that is the order in which they were presented. The system first does the equivalent of a BLDL for WOMBAT. As with any concatenation, the only one that will be found is the first one, which is the one that corresponds to FRED, and the BLDL will indicate that WOMBAT is an alias of FRED. If the module is reentrant, the system will then see that FRED is already loaded, will create a CDE for WOMBAT and will chain the WOMBAT CDE after the CDE for FRED. It will create an LLE for WOMBAT which will point to the CDE for WOMBAT. More interest is the case where nothing is loaded yet. The BLDL for WOMBAT is done. It indicates this is an alias of FRED A CDE is built for WOMBAT A BLDL for FRED is done. A CDE is built for FRED. FRED is loaded, FRED's CDE's use count is 1 . An LLE is built for WOMBAT. Its load count is 1. Even more interesting is the case where data set 1 has "FRED" and data set 2 has "FRED" with alias "WOMBAT". This too is a "do not do this" situation. Peter Relson z/OS Core Technology Design ---------------------------------------------------------------------- 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
