X-Original-To: pgsql-admin-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 5B7915281E for ; Wed, 15 Jun 2005 13:37:43 -0300 (ADT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 43113-02 for ; Wed, 15 Jun 2005 16:37:35 +0000 (GMT) Received: from chaiyo.com (unknown [203.148.250.248]) by svr1.postgresql.org (Postfix) with SMTP id 8EAC45280C for ; Wed, 15 Jun 2005 13:37:32 -0300 (ADT) Received: (qmail 16084 invoked by alias); 15 Jun 2005 16:38:40 -0000 Date: 15 Jun 2005 16:38:40 -0000 Message-ID: <20050615163840.16083.qmail@chaiyo.com> To: Tom Lane Subject: Re: Blob error after backup and restore [database > 5.5 Gb.] MIME-Version: 1.0 From: "Amrit Angsusingh" Cc: postgresql admin Content-Type: text/plain; charset=TIS-620; format=flowed Content-Transfer-Encoding: 8bit X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=1.382 tagged_above=0 required=5 tests=AWL, DNS_FROM_RFC_ABUSE, DNS_FROM_RFC_POST X-Spam-Level: * X-Archive-Number: 200506/101 X-Sequence-Number: 17856 The test database is on one of the fedora 3 server and postgresql 7.4.7 [rpm installed] . I back it up with the command pg_dump -Ft - b test > test.tar And once the process finished I transfer it to another server via nfs and restore it with pg_restore -d test test.tar I use this process for more than 1 1/2 year without any problem include the blob , but just for a couple of weeks ago the backup database has no blob and I cannot create and save a new blob data . Is that because the database is quite big [5.5 Gb] or not? Amrit Thailand "Amrit Angsusingh" writes: > after I try to restore it in another server which has the same environment [via nfs] and use the command > pg_restore -d test test.tar > The database "test" after restoration has the same data but no blob > found . Another thought here: what exactly do you mean by "no blob found"? The current implementation does not try to restore blobs with the same OIDs they had in the original database --- rather, it loads them up with whatever OIDs the new backend cares to assign, and then tries to fix up database columns of OID and LO types to reference the new OIDs instead of the old ones. This is obviously pretty fragile, and we are thinking about changing it for 8.1, but I wonder if you are getting bit because you are doing something that pg_restore isn't expecting. regards, tom lane ---------------------------(end of broadcast)--------------------------- TIP 9: the planner will ignore your desire to choose an index scan if your joining column's datatypes do not match ---------------------------------------------------------------------------------- http://www.chaiyo.com ไชโยเมล์ - ฟรีอีเมล์รูปแบบใหม่ สดใสกว่าเดิม...