In a message dated 10/26/2006 9:00:38 A.M. Central Daylight Time, [EMAIL PROTECTED] writes:
>That would require that the control unit generate a "state change" >asynch interrupt when each flashcopy relationship is finished. I don't >believe it does so. Some doc that I have suggests that it does generate >a state change when ALL active relationships on a given disk have >completed but I don't believe that z/OS does anything with that info. All my experiments using GTF to trace PPRC and Flashcopy I/Os were done several years ago when control units would only support one such relationship per disk. Back then each target disk's becoming synchronized with its source caused one state change interrupt, which was followed immediately by AOM's I/O chain (Read Device Characteristics, Read Configuration Data, and Sense Subsystem Status, but probably not in that order) to read the new state. I also observed that the DCE's contents changed after this read status I/O ended. This is the same chain that is executed during VARY ONLINE processing. I have no similar experience with the latest and greatest controllers that allow multiple FlashCopy relationships per device. Bill Fairchild "Facts are the enemy of truth." [Don Quixote in Man of La Mancha] ---------------------------------------------------------------------- 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
