I wrote:
> Andres Freund <andres@anarazel.de> writes:
>> It'd not be unreasonble to check pg_control first, and only after that
>> indicates readyness check via the protocol.
> Hm, that's a thought. The problem here isn't the frequency of checks,
> but the log spam.
Actually, that wouldn't help much as things stand, because you can't
tell from pg_control whether hot standby is active. Assuming that
we want "pg_ctl start" to come back as soon as connections are allowed,
it'd have to start probing when it sees DB_IN_ARCHIVE_RECOVERY, which
means Jeff still has a problem with long recovery sessions.
We could maybe address that by changing the set of states in pg_control
(or perhaps simpler, adding a "hot standby active" flag there). That
might have wider consequences than we really want to deal with post-beta1
though.
regards, tom lane