Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.96) (envelope-from ) id 1vv3jg-001K44-0E for pgsql-hackers@arkaria.postgresql.org; Wed, 25 Feb 2026 01:30:44 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.96) (envelope-from ) id 1vv3je-004BSY-1Y for pgsql-hackers@arkaria.postgresql.org; Wed, 25 Feb 2026 01:30:42 +0000 Received: from magus.postgresql.org ([2a02:c0:301:0:ffff::29]) by malur.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.96) (envelope-from ) id 1vv3je-004BSQ-0d for pgsql-hackers@lists.postgresql.org; Wed, 25 Feb 2026 01:30:42 +0000 Received: from sss.pgh.pa.us ([68.162.161.243]) by magus.postgresql.org with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.98.2) (envelope-from ) id 1vv3jb-000000018xJ-0GL3 for pgsql-hackers@lists.postgresql.org; Wed, 25 Feb 2026 01:30:41 +0000 Received: from sss1.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.pgh.pa.us (8.15.2/8.15.2) with ESMTP id 61P1UTjV082508; Tue, 24 Feb 2026 20:30:29 -0500 From: Tom Lane To: Jacob Champion cc: Andrew Dunstan , Jelte Fennema-Nio , pgsql-hackers@lists.postgresql.org Subject: Re: pgsql: libpq: Grease the protocol by default In-reply-to: References: <3626145.1771885089@sss.pgh.pa.us> <3635497.1771889682@sss.pgh.pa.us> <3797860.1771893905@sss.pgh.pa.us> <3864574.1771898936@sss.pgh.pa.us> Comments: In-reply-to Jacob Champion message dated "Tue, 24 Feb 2026 14:59:04 -0800" MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <82506.1771983029.1@sss.pgh.pa.us> Date: Tue, 24 Feb 2026 20:30:29 -0500 Message-ID: <82507.1771983029@sss.pgh.pa.us> List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Archived-At: Precedence: bulk Jacob Champion writes: > Oh, the envvar is clever. You'll probably want to do that only for the > pg_dumpall invocation now that pg_upgrade is patched, though, so we > don't cover up regressions. I can confirm clean x-version tests on all branches with git tip and this: --- TestUpgradeXversion.pm.orig 2025-11-25 07:47:25.000000000 -0500 +++ TestUpgradeXversion.pm 2026-02-24 18:50:29.487530840 -0500 @@ -485,10 +485,14 @@ sub test_upgrade ## no critic (Subrou # use the NEW pg_dumpall so we're comparing apples with apples. setinstenv($self, "$installdir", $save_env); + local $ENV{PGMAXPROTOCOLVERSION} = + ($oversion le 'REL9_2_STABLE') ? "3.0" : "latest"; + system( qq{"$installdir/bin/pg_dumpall" $dump_opts -p $sport -f } . qq{"$upgrade_loc/origin-$oversion.sql" } . qq{> "$upgrade_loc/$oversion-dump1.log" 2>&1}); return if $?; + delete $ENV{PGMAXPROTOCOLVERSION}; setinstenv($self, "$other_branch/inst", $save_env); system( qq{"$other_branch/inst/bin/pg_ctl" -D } which is Andrew's patch but with the envvar dropped as soon as possible. regards, tom lane