Re: pgsql: libpq: Grease the protocol by default

From: Jelte Fennema-Nio <postgres(at)jeltef(dot)nl>
To: Jacob Champion <jacob(dot)champion(at)enterprisedb(dot)com>
Cc: Jacob Champion <jchampion(at)postgresql(dot)org>, pgsql-committers(at)lists(dot)postgresql(dot)org
Subject: Re: pgsql: libpq: Grease the protocol by default
Date: 2026-02-23 22:24:34
Message-ID: CAGECzQSCd+zCc4p+wqC9dA6yw86dXdt=sd+i3toB66TAz60WvQ@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-committers pgsql-hackers

On Mon, 23 Feb 2026 at 22:17, Jacob Champion
<jacob(dot)champion(at)enterprisedb(dot)com> wrote:
> So while this is kind of a nice result, in that it's correctly
> diagnosing the problem, I need to add max_protocol_version=3.0 to the
> connection strings for these upgrades during beta.

Two thoughts:
1. I think we might want the "This indicates a bug in either the
server being contacted..." note to be added to the error even if you
use 3.2/3.3/3.X. It seems pretty helpful in general. It would need a
bit fancier detection though, with strings to search for based on the
actual requested version.
2. Why only force max_protocol_version=3.0 for beta? It sounds like
this would also be an issue once we eventually bump the default
version.

In response to

Browse pgsql-committers by date

  From Date Subject
Next Message Jacob Champion 2026-02-23 22:51:46 Re: pgsql: libpq: Grease the protocol by default
Previous Message Tom Lane 2026-02-23 22:18:09 Re: pgsql: libpq: Grease the protocol by default

Browse pgsql-hackers by date

  From Date Subject
Next Message Andres Freund 2026-02-23 22:27:59 Re: Reduce timing overhead of EXPLAIN ANALYZE using rdtsc?
Previous Message Tom Lane 2026-02-23 22:18:09 Re: pgsql: libpq: Grease the protocol by default