> Have the UTL exit schedule an IRB to run on evry TCB you want stopped. > After all IRB's are scheduled, the UTL code waits on one or more ecb's. > There's one ecb for each task running you want stopped > > The IRB code posts it's UTL ecb indicating it has arrived. It then waits > on > a different ecb, call it the IRB ecb. > > After all the UTL ecb's have been posted, issue the wtor. Post the IRB > ecb's, then take whatever action in the UTL code you want based upon the > operator reply. > > After the IRB ecb is posted, the IRB code just exits. > > Seems like it oughta work.
It will work FSVO of "work" but the proposed solution has the same set of issues that STATUS does. You would be preventing execution of work within the address space. Most of the time that would be fine, but you just can't predict what you would be interrupting. If the interrupted work holds some resource that is needed by other work, then your exit can get the whole system wrapped around the axle so fast you may not have any idea what's actually causing it. FWIW my advice is just don't mess with it. It would be a bad idea to get real creative in your exit. If you decide you need to do something (change state, reset, cancel or whatever) just belch up a WTO from your exit and let automation handle 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
