Re: sslmode=secure by default (Re: Making sslrootcert=system work on Windows psql) - Mailing list pgsql-hackers

From Jacob Champion
Subject Re: sslmode=secure by default (Re: Making sslrootcert=system work on Windows psql)
Date
Msg-id CAOYmi+kqQAWL_0Pi87xA5qS84iHKxY4B-NCMhqF+ocaoS3R7DA@mail.gmail.com
Whole thread Raw
In response to Re: sslmode=secure by default (Re: Making sslrootcert=system work on Windows psql)  (Peter Eisentraut <peter@eisentraut.org>)
List pgsql-hackers
On Thu, Apr 24, 2025 at 5:00 AM Peter Eisentraut <peter@eisentraut.org> wrote:
> Another detail to think about is how this affects psql -h localhost.  In
> principle, this should require full SSL, but you're probably not going
> to have certificates that allow "localhost".  And connections to
> localhost are the default on Windows.  We could also switch the Windows
> default to Unix-domain sockets.  But there are probably still other
> reasons why connections to TCP/IP localhost are made.  Some things to
> think about.

Yeah, we pretty quickly get to the boring-but-hard part. Is there a
group of users we feel comfortable breaking? What ways is it
acceptable to break them? How hard should it be for them to unbreak
themselves once it happens?

It'd be kind of nice if there were a better way than environment
variables to configure defaults for the client. I've been looking at
openssl.cnf for the Windows certificate problem, and I wish we had
that knob available for conversations like this... If we had a global
client config, then we could declare that we're going to change the
defaults in that config far in advance, and anyone who absolutely
hates it can proceed to undo it globally and move on. The service file
is IMO not enough for this.

--Jacob



pgsql-hackers by date:

Previous
From: Jacob Champion
Date:
Subject: Re: sslmode=secure by default (Re: Making sslrootcert=system work on Windows psql)
Next
From: "David E. Wheeler"
Date:
Subject: Re: extension_control_path and "directory"