There's a recent CICS APAR PI82188 that showed up in my Red Alert feed. Evidently CICS was using STCKF to obtain a "unique" value as a Unit Of Work (UOW) identifier, and once in a while two STCKFs returned the same value, with Very Bad consequences. The fix is to change it (back, by the sound of things) to STCK, with all the cross-CPU expense that can cause. One imagines they'll eventually convert to STCKE, but that's not a drop-in change.
STCKF and friends have been discussed here and/or the assembler list in recent years, but I've always thought of STCKF clashes as a theoretical problem not likely to be encountered in one's lifetime. The APAR description suggests, without clearly saying so, that this has happened at multiple customer sites. If you're one in a million, and you live in a city of three million people... Tony H. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
