Sarathy, Thank you for your clear explanation. We'll look into this, but I do not believe that we can spend the time to fix this larger problem. I think others will have to carry the ball rest of the way; I hope someone will have the opportunity to do it soon. In the meantime, we will continue to use our patched version until something better comes along. I would like to continue to offer to test alternative fixes to the one we produced. We have a large (>17K line) multi-threaded perl application that I have had run for more than a week and used to spawn more than a hundred threads (at peak). It has a _lot_ of unnecessary locks to get around the bug that our patch fixes, but it still serves as a nice perl thread testbed. I will go back to removing all of those locks now, and doing further testing. Sincerely, Rob Gurusamy Sarathy wrote: > [A detailed description of the problem.] > > Sarathy > gsar@ActiveState.com -- Dr. Robert K. Cunningham Information System Technology Group rkc@ll.mit.edu MIT Lincoln Laboratory *** My comments, my opinions: my responsibility. PGP key available from http://pgpkeys.mit.edu:11371Thread Previous | Thread Next