Received: from sss.sss.pgh.pa.us (sss.pgh.pa.us [206.210.65.6]) by hub.org (8.9.3/8.9.3) with ESMTP id MAA61963 for ; Mon, 10 May 1999 12:59:27 -0400 (EDT) (envelope-from tgl@sss.pgh.pa.us) Received: from sss.sss.pgh.pa.us (localhost [127.0.0.1]) by sss.sss.pgh.pa.us (8.9.1/8.9.1) with ESMTP id MAA07803; Mon, 10 May 1999 12:58:07 -0400 (EDT) To: "Frank Morton" cc: pgsql-sql@postgreSQL.org Subject: Re: [SQL] Slow Inserts Leads To Unable To Dump In-reply-to: Your message of Mon, 10 May 1999 08:17:05 -0500 <041c01be9ae7$668ea5a0$8355e5ce@base2inc.com> Date: Mon, 10 May 1999 12:58:07 -0400 Message-ID: <7801.926355487@sss.pgh.pa.us> From: Tom Lane "Frank Morton" writes: > Now that that is done, I thought I should dump the database > before doing any more to save those 7 days if I mess up. > After processing 18 HOURS doing the "pg_dump -d", it ran > out of memory and quit. Are there tools to do this differently, Yes: pg_dump without -d. COPY should work fine. This does suggest an item for the to-do list: making pg_dump use DECLARE CURSOR and FETCH rather than a plain SELECT to pull the contents of tables in -d mode. However, if your table is so large that it doesn't fit in memory inside pg_dump, then I'd have to say that you wouldn't want to reload it from a series of INSERT statements anyway. So I'm not very excited about this... It does seem that your installation is remarkably slow; other people are working with tables exceeding 2Gb and don't seem unhappy about the speed. Perhaps you need to check to see if the thing is thrashing; how much real memory do you have, and what -B setting are you using? regards, tom lane