Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtp (Exim 4.84_2) (envelope-from ) id 1eIcFJ-0000YZ-19 for pgsql-admin@arkaria.postgresql.org; Sat, 25 Nov 2017 15:19:57 +0000 Received: from localhost ([127.0.0.1] helo=malur.postgresql.org) by malur.postgresql.org with esmtp (Exim 4.84_2) (envelope-from ) id 1eIcFI-0007sU-8o for pgsql-admin@arkaria.postgresql.org; Sat, 25 Nov 2017 15:19:56 +0000 Received: from makus.postgresql.org ([2001:4800:1501:1::229]) by malur.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA384:256) (Exim 4.84_2) (envelope-from ) id 1eIcFG-0007Os-7s for pgsql-admin@lists.postgresql.org; Sat, 25 Nov 2017 15:19:54 +0000 Received: from sss.pgh.pa.us ([66.207.139.130]) by makus.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA384:256) (Exim 4.89) (envelope-from ) id 1eIcFD-0007YF-Rg for pgsql-admin@postgresql.org; Sat, 25 Nov 2017 15:19:52 +0000 Received: from sss1.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.pgh.pa.us (8.14.4/8.14.4) with ESMTP id vAPFJa34000777; Sat, 25 Nov 2017 10:19:36 -0500 From: Tom Lane To: Debraj Manna cc: Mark Kirkwood , Alvaro Aguayo Garcia-Rada , pgsql-admin Subject: Re: Faster way of upgrading postgresql to 10 from 9.5 In-reply-to: References: <4d397bf0-14c3-2bf8-bee6-c023a654b7ef@catalyst.net.nz> Comments: In-reply-to Debraj Manna message dated "Sat, 25 Nov 2017 11:48:27 +0530" MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-ID: <775.1511623176.1@sss.pgh.pa.us> Date: Sat, 25 Nov 2017 10:19:36 -0500 Message-ID: <776.1511623176@sss.pgh.pa.us> List-Id: List-Help: List-Subscribe: List-Post: List-Owner: List-Archive: Precedence: bulk Debraj Manna writes: > Is this time expected? I mean 17 minutes for 490 GB. I could not find any > benchmark for pg_upgrade. It seems awfully slow, if you're using -k. That switch should remove any dependency on the data size, but what remains is the number of SQL objects (databases, tables, functions, etc). If you've got tens or hundreds of thousands of those, maybe this would be a reasonable time. As Alvaro said, it's possible that you're hitting some fixable bottleneck; but to identify what that is we'd need some profiling information. regards, tom lane