"Kevin Grittner" <Kevin.Grittner@wicourts.gov> writes:
> Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> [ thinks... ] Maybe we could have the postmaster generate a random
>> number at start and include that in both the postmaster.ports file
>> and its pg_ping responses.
> Unless two postmasters could open the same server socket within a
> microsecond of one another, a timestamp value captured on opening the
> server socket seems even better than a random number.
Well, that raises the question of whether postmaster uptime could be
considered security-sensitive info. I'd still rather use a random
number.
regards, tom lane