Jim Mulder quoth; > I make no statements about the safety of using FORCE. The ramifications > of any asynchronous abnormal termination depend on what was going on when > the termination occurred. That is true FORCE, CANCEL , DETACH, etc.
As Jim says, there are lots of "interesting" cases that can sometimes arise in asynchronous termination. What I used to see -very- often was that asynchronous abends nailed the mainline of, or a recovery routine of, or end-of-task exits of certain BCP components (particularly VSM) and since those components were apparently not intended to be summarily terminated, "unpredictable" results occurred. There are also situations where, since the advent of parallel detach, circa ESA V4, tasks can terminate out of order with respect to the order expected by the application running in the address space. This may result in one or more ECB's disappearing and, due to a decades old bug in POST, any task waiting on such an ECB cannot be taken out of wait state (basically POST gets a S0C4-10 or 11 trying to reference the ECB to take it out of wait status and it surfaces that abend to the POST caller, rather than asynchronously abending the waiting task) The net of all that is that the orderly task termination sequence gets hosed and it is entirely possible the address space will hang in a crippled state until you come along with a FORCE. Is it an APAR-able defect? Maybe, but whose defect is it? CC ---------------------------------------------------------------------- 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
