Bruce Momjian <pgman@candle.pha.pa.us> writes:
>> How about calling pg_ctl from pg_upgrade to stop the postmaster ?
> Great idea. We never had an automated way to do that before.
I'd just as soon *not* have pg_upgrade assuming that it knows how
I like my postmaster started. If pg_ctl will work to stop a postmaster
whether or not it was started with pg_ctl, then it'd probably be a good
idea to use pg_ctl to ensure that there is no postmaster running in the
target database. I'd suggest not trying to restart the postmaster
automatically afterwards, though. Too much site dependency in that.
regards, tom lane