Pommier, Rex wrote, on IBM-MAIN, in response to "What is 'Easter Egging'?":
>Easter Egging in the more general sense is trying to fix something by randomly replacing elements when you don't have any idea what to fix and hoping you get the right thing. Based on the custom in various parts of the world of hunting for Easter Eggs. >Also known as shotgun debugging. That's an accurate description, but has a negative connotation that I'd defend against. When you've exhausted the normal approaches to something, changing things that should have an effect-just not one that's expected to fix the problem-is often illuminating. As in this example. We had started with all the right approaches; they were just fruitless because the display was lying to us. Putting a WTO in a program when you can't tell where it's ABENDing (and don't have proper tools to debug it otherwise, obviously) is even an example: "Ah, ok, we put out that message before the S0C1, so we did get THAT far." Elardus suggested: > That class was probably a dynamic class and Vanguard was perhaps not updated/upgraded to pick it up? Well, a followup from Vanguard suggests: "it is possible that someone has defined the CADB2 entity in the static Class Descriptor table as well as the Dynamic CDT and the Dynamic CDT always gets checked first." Not sure whether this addresses that or not. And no, the POSIT wasn't in the IBM range. I'm going to see if the customer has time to run LISTCDT. .phsiii ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
