>Also, it might be worth enabling log_lock_waits to see if the slow
>notifies are due to having to wait on some lock or other.
Turning on log_lock_waits shows that there is a lot of waiting for
locks on the pg_listener table ala:
process 22791 still waiting for ExclusiveLock on relation 2614 of
database 16387 after 992.397 ms
...
process 22791 acquired ExclusiveLock on relation 2614 of database
16387 after 1433.152 ms
deadlock_timeout is left at the default 1s setting.
Though these are being generated during application activity -
running the simulation script does produce 300ms - 600ms NOTIFY
statements but doesn't (at the moment) trigger a lock_wait log entry.