Jay Flattery <jaycode@xxxxxxxxxxxxxx> writes: > Actually, if I just dump the example into a file, compile and install it as a > function, it works. > Only fails when I add the code (without changing anything!) to existing source > where I'd like to use it. Hmph, are you trying to incorporate this code into the core backend? It's depending on being in a dynamic library, because otherwise that _PG_init function won't get called. (Although that still seems like it doesn't explain the symptom ... should crash sooner if the hashtable's not been created.) FWIW, that backtrace looks like it's from the postmaster process, not from a crashing backend. You need to make sure you attach gdb to the right process ... regards, tom lane -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
