Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1l5Wpr-0001PN-7w for pgsql-bugs@arkaria.postgresql.org; Fri, 29 Jan 2021 16:41:27 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.92) (envelope-from ) id 1l5WpJ-0003oK-9n for pgsql-bugs@arkaria.postgresql.org; Fri, 29 Jan 2021 16:40:53 +0000 Received: from magus.postgresql.org ([2a02:c0:301:0:ffff::29]) by malur.postgresql.org with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1l5WpJ-0003oC-1c for pgsql-bugs@lists.postgresql.org; Fri, 29 Jan 2021 16:40:53 +0000 Received: from mout.gmx.net ([212.227.15.19]) by magus.postgresql.org with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1l5WpG-0006oH-Ud for pgsql-bugs@lists.postgresql.org; Fri, 29 Jan 2021 16:40:52 +0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=gmx.net; s=badeba3b8450; t=1611938448; bh=K8D8WH95bCK7y22hsgrY0+YbQRfu5GhpvZpsaRgdBvY=; h=X-UI-Sender-Class:From:To:Cc:Subject:Date:In-Reply-To:References; b=g+suSUBK4jgnwr71Br1WneCfcTe6WVuBzXEytDoNlU2m4B2HeqVJ1w/2+jrP7JrFU tbgGNUajaq9qcVAto+STz0fTJgIXAPsyMUS/bcn45ysJURNM9tBTx4jUxdGjEANpsg adMUREDWfb1k3RXKe2s9WG/Jgo9fUS+oZemx/xUc= X-UI-Sender-Class: 01bb95c1-4bf8-414a-932a-4f6e2808ef9c Received: from wolfclan.ang.de ([87.191.134.178]) by mail.gmx.net (mrgmx005 [212.227.17.190]) with ESMTPSA (Nemesis) id 1MFbW0-1lBWJS0qh4-00HAag; Fri, 29 Jan 2021 17:40:48 +0100 From: GMX LINREG To: Tom Lane Cc: Bruce Momjian , pgsql-bugs@lists.postgresql.org Subject: Re: BUG #16843: pg_upgrade from 12.5 to 13.1 with extension plperlu failed Date: Fri, 29 Jan 2021 17:40:46 +0100 Message-ID: <7912997.NyiUUSuA9g@wolfclan.ang.de> In-Reply-To: <2780634.1611932694@sss.pgh.pa.us> References: <16843-174d3e5b66729a60@postgresql.org> <7226587.EvYhyI6sBW@wolfclan.ang.de> <2780634.1611932694@sss.pgh.pa.us> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="iso-8859-1" X-Provags-ID: V03:K1:/0oBJlgwVLtPuy62sXfdYf3U6IwYmcfliH6P5Fh0OWozxdk+Kuy +mKbYXrTB3gCSABSFt3Acu/fcbf8HDCy6Fh+Cbez2TgkPpA9ZtKiZd+BMUtbAySDt9K8gUH LT/uWmDs2mxEfCmloe1NahyI1ESoMsvJ4ojSV2PtbMQEgwZaEhR6JDcmoYcBCECiew5kDda ayDojSuDr0ZkjGsNRVrLQ== X-Spam-Flag: NO X-UI-Out-Filterresults: notjunk:1;V03:K0:enpu3Sw1jHA=:kFRjM/w6XVxksViz/WvAwy PUcj8yirHMr2SUZ/qDbmaib4lRT9WFUkpc6zWboNsbiidaJT+WYxN2KOT2zE2iYJJ2ytqPekn OeaiJlPRngW/v8gzFn0Up0iS2ReRy2bRaIJQNCjuTz9uMm3HIm0GtSDDrwEKiR0YwuES90OXS pBv/IFsfiTBdyx8cDRvQOvT1JA+eok15zz0qpiOPUdKbGG7+izwn/P8eWMvyy1CZl6QECBNAT i56l8Uib/AUzoP/2v8fjiJ97kKvbmDz8UozGl5ez57WSxdR8vJPZY+AFYbxg+cUaPS5+gQVtN SCEVdsnNi1IZvp0VtI0ZzJENX02R+71pHdQ5mHFcHvB09XRVM8xjQeqpTVD8b+CTjI8SDsmBQ uGxSYNU/C2h0v50/hydTEAuvtjNp+0j5SwQcIGZnzbV9a435VIv0ojiYQOKZf6vse4OPyCY0n lyakuGhdKpJHlPjj6IyTnkWofVQmBnFl/3ZomgLKlz1j6EuWJXB4ZdH62Y8ozdfyHkm4lNYPW qSPy5e0W8Ib7SnrGGBhkdX7oUsrRyJZWJyqHSw9Vq04C+GBs/cxW8go6iujFYegDVggYUNSCz H3X+6UiiP0dnY84jVGaJ6RxibewQXV27DgtUCR97wz9aG3QqG/M5V2WAkR6QNzVLAbhhcmhFf Wpw3Zi0AVMmBmbJ+Z21oG79dT6G5SsZaQ0kwLGsb4Ue0U2MJ7B00H1B99ZWxqxO1n2PpaMhwy ZkIcZdXnjAh7hfY6cODPdTdpGlsr3Ynu4536mWsKszSGflLUFzXSTPZGD/FDjjFySvSwX7/bp HJM2qzZQvyDwWda9h02kqrpFkLvXKNmyNiM6YgmEJjjs880XjEniULYkDN9e1ZAZ8w93P1j1u GrtMBUpZjTjK5LPq7a5g== List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Precedence: bulk Am Freitag, 29. Januar 2021, 16:04:54 CET schrieb Tom Lane: > GMX Steffen writes: > > OS Packages: > > *postgresql*-llvmjit-13-lp152.3.1.noarch > > *postgresql*12-pldebugger-1.0+git13.ddbce7b-lp152.1.4.x86_64 > > *postgresql*12-llvmjit-12.5-lp152.34.1.x86_64 > > *postgresql*-13-lp152.3.1.noarch > > *postgresql*12-docs-12.5-lp152.34.1.noarch > > *postgresql*12-contrib-12.5-lp152.34.1.x86_64 > > *postgresql*13-plperl-13.1-lp152.16.1.x86_64 > > *postgresql*12-server-12.5-lp152.34.1.x86_64 > > *postgresql*-docs-13-lp152.3.1.noarch > > *postgresql*12-12.5-lp152.34.1.x86_64 > > *postgresql*12-pg_qualstats-2.0.2-lp152.2.3.x86_64 > > *postgresql*12-ip4r-2.4.1+git1.5f9ce88-lp152.8.2.x86_64 > > *postgresql*-server-13-lp152.3.1.noarch > > *postgresql*12-pg_qualstats-llvmjit-2.0.2-lp152.2.3.x86_64 > > *postgresql*13-docs-13.1-lp152.16.1.noarch > > *postgresql*13-server-13.1-lp152.16.1.x86_64 > > *postgresql*-contrib-13-lp152.3.1.noarch > > *postgresql*12-ip4r-llvmjit-2.4.1+git1.5f9ce88-lp152.8.2.x86_64 > > *postgresql*13-13.1-lp152.16.1.x86_64 > > *postgresql*13-llvmjit-13.1-lp152.16.1.x86_64 > > *postgresql*-plperl-13-lp152.3.1.noarch > > *postgresql*12-pldebugger-llvmjit-1.0+git13.ddbce7b-lp152.1.4.x86_64 > > *postgresql*13-contrib-13.1-lp152.16.1.x86_64 >=20 > This ... looks like a bit of a mess. You evidently have three independent > sets of Postgres packages, but surely there should only be two. And > why do they all have "lp152" in the version? (And why are there only > two versions of plperl? And why are some of these "noarch"? That > would make sense for the docs subpackage, but not much else.) >=20 > The pg_restore trace gives us no more info than we had before. But > given this package list, I am suspecting some confusion over which > version of plperl.so should get loaded. You might try looking into > the destination postmaster's log file (not pg_upgrade's user-visible > output) to see if there is any low-level message from the dynamic > loader emitted just before it complains about plperlu not existing. >=20 > regards, tom lane Hello, =2D packages like "postgresql*-contrib-13-lp152.3.1.noarch" are meta packag= es (noarch) and depend to actual installed version. in this case version 13= =2E=20 =2D lp152 is a label for opensuse leap 15.2 =2D when run version 12 it use /usr/lib/postgresql12/lib64/plperl.so for version 13 it's use /usr/lib/postgresql13/lib64/plperl.so =2D there is no low-level message from the dynamic loader. the extension ca= n created everytime. And exist in the upgraded database=20 i hope you could read my last email: i resend three times :( There you can see this:=20 psql (13.1) Geben Sie =BBhelp=AB f=FCr Hilfe ein. botdb=3D# \dL Liste der Sprachen Name | Eigent=FCmer | Vertraut | Beschreibung =20 =2D--------+------------+----------+------------------------------ plpgsql | postgres | t | PL/pgSQL procedural language (1 Zeile) =3D=3D> NO LANGUAGE PLPERLU! "CREATE LANGUAGE plperlu" =3D=3D> create ONLY = an extension!=20 botdb=3D# \dx Liste der installierten Erweiterungen Name | Version | Schema | Beschreibung = =20 =2D-------------+---------+------------+-----------------------------------= =2D-------------- hstore | 1.5 | pg_catalog | data type for storing sets of (key, = value) pairs pg_cron | 4.0.4 | public | Cron-like Job Scheduler for PostgreS= QL pg_http | 1.4.1 | public | A web service for PostgreSQL pg_mqtt | 1.1 | public | A mqtt client daemon for PostgreSQL pg_reporting | 1.1 | public | A reporting extension for PostgreSQL plperlu | 1.0 | pg_catalog | PL/PerlU untrusted procedural langua= ge <=3D=3D=3D=3D plpgsql | 1.0 | pg_catalog | PL/pgSQL procedural language (7 Zeilen) =3D=3D> EXTENSION plperlu ALREADY CREATED AND EXIST=20 The command "create language" create NO plperlu language object. ONLY a ext= ension object. Correct? Therefor it must fail to add a language to an extension! And i think the "c= reate extension" run fine, so that the command alter extension / add langua= ge combination is only for older postgresql version okay. it is not needed = any more. The failed upgraded database has the extension plperlu installed! So this i= s fine too. best regards thomas steffen