Re: [COMMITTERS] pgsql: Improve postmaster's logging of listen socket creation.

Stephen Frost <sfrost@snowman.net>

From: Stephen Frost <sfrost@snowman.net>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Andres Freund <andres@anarazel.de>, Robert Haas <robertmhaas@gmail.com>, "pgsql-hackers@postgresql.org" <pgsql-hackers@postgresql.org>
Date: 2017-03-14T11:37:01Z
Lists: pgsql-hackers
* Tom Lane (tgl@sss.pgh.pa.us) wrote:
> Andres Freund <andres@anarazel.de> writes:
> > I don't mind the new output, but I kinda wonder whether it's a good idea
> > to include the '.s.PGSQL.5432' bit in the host and/or whether we
> > shouldn't include the port in the TCP cases as well
> 
> Yeah, I've been thinking that maybe it should look like
> 
> 2017-03-13 10:08:49.399 EDT [90059] LOG:  listening on IPv6 address "::1", port 5432
> 2017-03-13 10:08:49.399 EDT [90059] LOG:  listening on IPv4 address "127.0.0.1", port 5432
> 2017-03-13 10:08:49.400 EDT [90059] LOG:  listening on Unix address "/tmp/.s.PGSQL.5432"
> 
> It would take a couple more lines of code to make that happen, but
> it would future-proof the messages against the day we decide to
> allow one server to respond to more than one port number ...

I certainly agree with adding the port for TCP.  I also agree with
Andres' point about the unix socket, though I'm tempted to suggest that
we should just teach libpq to understand a straight unix socket being
passed for host/-h rather than change what the server reports here..

Thanks!

Stephen

Commits

  1. Include port number when logging successful binding to a TCP port.

  2. Improve postmaster's logging of listen socket creation.