Error -3106

Good day, good people:

I am running on a Pyramid (AT&T 7080) using RDBMS 6.0.36.0.1 and Pro*C 1.4.8.2.2 (UNIX environment). I have several different kinds of processes written in Pro*C that do pretty normal stuff (inserts, selects, updates, some Dynamic SQL). I am getting error -3106 ("fatal two-task communication protocol error") at various points in the processes. One process will run fine for days, get the error, and then be fine again. Another process will run fine, get the error once, and then always get the error after that. It is happening quite frequently in our Development site and happens at least once a day in our Production site. Our RDBMS and our application are running on the same machine. I cannot discern any pattern to the error occurrences.

Has anyone else ever seen this error? Does anyone know under which circumstances this error occurs?

Thanks,
Marianne Received on Thu Mar 04 1993 - 16:32:24 CET