I can certainly fix COMMAND, but as it has been this way for over twenty years, I shall do so only if IBM will pay me handsomely to do it. And I'd expect that one would find frost in a particular place first.
I'm sure there a lots of other stages that expose session resources, which would all need serialisation; they may even need to hold a lock for the duration. Also bear in mind that anything else in MT would also need to respect my own locking conventions, clearly not realistic. j. On 28 August 2010 19:53, Rich Greenberg <[email protected]> wrote: > On: Sat, Aug 28, 2010 at 06:06:47AM +0200,John P. Hartmann Wrote: > > } Running multiple CMS commands concurrently is asking for trouble; and > } running multiple COMMAND stages concurrently even more so, as the > } magic and mirrors that make it happen is in parts of CMS that is not > } re-entrant. > > John, > Not a real fix, but one possible band aid might be that at the beginning > of stages like COMMAND, you check for running MT, and if so, check a > lock. If not set, set it, do the COMMAND, then unset. If already set, > wait a bit and try again. > > -- > Rich Greenberg Sarasota, FL, USA richgr atsign panix.com + 1 941 378 2097 > Eastern time. N6LRT I speak for myself & my dogs only. VM'er since CP-67 > Canines: Val, Red, Shasta, Zero & Casey (At the bridge) Owner:Chinook-L > Canines: Red & Cinnar (Siberians) Retired at the beach Asst Owner:Sibernet-L >
