The problem only occurs when the value of MSH-10 is the same in two transactions sent in the same batch. When I send out a batch consisting only of the original transaction plus the newly-created one (with MSH-10 incremented by 1 in the xlate), it works fine. However, if there are additional transactions in the batch, and if one of them duplicates the value of MSH-10 in the new transaction, everything stops flowing and inbound replies (of the new transaction) are generated and placed in pending (state 11) every 60 seconds (the timeout value). I don’t know Cloverleaf well enough to understand what’s going on, but a TCL that increments MSH-10 for every transaction (and provides an accurate message count in BTS-1) would probably solve the problem.
