Bruce Momjian <bruce@momjian.us> writes:
> Agreed. So how do we pass that info to libpq without exceeding the
> value of fixing this problem? Should we parse pg_controldata output?
> pg_upgrade could use machine-readable output from that too.
pg_controldata seems 100% unrelated to this problem. You cannot even
tell if the postmaster is alive just by inspecting pg_control.
>> What we actually want here, and don't have, is the fabled pg_ping
>> protocol...
> Well, we are basically figuring how to implement that with this fix,
> whether it is part of pg_ctl or a separate binary.
Possibly the cleanest fix is to implement pg_ping as a libpq function.
You do have to distinguish connection failures (ie connection refused)
from errors that came back from the postmaster, and the easiest place to
be doing that is inside libpq.
regards, tom lane