On Mon, Feb 23, 2026 at 4:45 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Jacob Champion <jacob.champion@enterprisedb.com> writes:
> > Something like the attached (tested only against 9.2 so far)? I would
> > plan to backpatch after feature freeze is lifted.
(er, should have been "release freeze", though we're clearly in no hurry)
> Also: I was initially baffled why you thought this needs
> back-patching, but I guess you have one eye on packagers like
> Debian who think they can make older versions use newer libpq.so.
Right.
> It'd be good to spell out that reasoning in the commit message.
Okay, will do.
> It's not obvious to me where the problem lies. I can replicate
> the failure by trying to use HEAD's psql to connect to a 9.2
> server, but adding -d "max_protocol_version=3.0" makes psql happy,
> so why not pg_dumpall?
Hmmm, looks like the -dump1.log output is actually from *before*
pg_upgrade actually runs:
https://github.com/PGBuildFarm/client-code/blob/28d7e945cc2a27fecdf4cc685782821ca504db5d/PGBuild/Modules/TestUpgradeXversion.pm#L514
So that will still need to be modified in the buildfarm client,
independently of my patch. (I reproduced the error with a bare
pg_upgrade invocation and didn't think to look closer; sorry for the
confusion.)
--Jacob
