Received: from localhost (unknown [200.46.204.191]) by postgresql.org (Postfix) with ESMTP id 574862E0284 for ; Fri, 25 Jan 2008 16:04:46 -0400 (AST) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.191]) (amavisd-maia, port 10024) with ESMTP id 29798-08 for ; Fri, 25 Jan 2008 16:04:39 -0400 (AST) Received: from isabelle.production.illumen.org (mail.illumen.org [216.147.207.247]) by postgresql.org (Postfix) with ESMTP id 41E7A2E00B5 for ; Fri, 25 Jan 2008 16:04:39 -0400 (AST) Received: from localhost (localhost.localdomain [127.0.0.1]) by isabelle.production.illumen.org (Postfix) with ESMTP id 2EC30310427 for ; Fri, 25 Jan 2008 15:04:38 -0500 (EST) X-Virus-Scanned: amavisd-new at Received: from isabelle.production.illumen.org ([127.0.0.1]) by localhost (isabelle.production.illumen.org [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id CxvkmFbM5IO3 for ; Fri, 25 Jan 2008 15:04:32 -0500 (EST) Received: from GatewayAMD64 (dsl093-111-018.wdc2.dsl.speakeasy.net [66.93.111.18]) by isabelle.production.illumen.org (Postfix) with ESMTP id F0C59310421 for ; Fri, 25 Jan 2008 15:04:31 -0500 (EST) From: "Adam Talaat" To: Subject: Data file compatibility between 32-bit and 64-bit Date: Fri, 25 Jan 2008 15:05:38 -0500 Message-ID: MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_002B_01C85F63.BEAABA40" X-Mailer: Microsoft Office Outlook 11 Thread-Index: AchfjacpLVFtePKYRw2rPd6dP1rfsQ== X-MimeOLE: Produced By Microsoft MimeOLE V6.0.6000.16545 X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0.002 tagged_above=0 required=5 tests=BAYES_50=0.001, HTML_MESSAGE=0.001 X-Spam-Level: X-Archive-Number: 200801/250 X-Sequence-Number: 28024 This is a multi-part message in MIME format. ------=_NextPart_000_002B_01C85F63.BEAABA40 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hello, Are the database files created by 32-bit PostgreSQL identical to the files created by 64-bit PostgreSQL? The database files would reside on a SAN or DAS device external to the servers. If the server running a 64-bit installation of PostgreSQL were to fail, would we be able to bring a server running a 32-bit installation online and run it against the original database files? We would absolutely not be simultaneously running two instances of the database against the same database files. My background is with MS SQL Server, where a database can be "reattached" to a different instance of the software (although not between 32-bit and 64-bit versions). We would be running PostgreSQL on RHEL/CentOS 5. I realize having a separate server is not a sufficient disaster recovery strategy. We do regular database backups to tape in the event of data corruption or catastrophic failure. Replication with Slony-I is also an option, however a complete hot or warm spare configuration would require additional storage space. Is it possible to replicate from 64-bit to 32-bit? If you haven't already guessed, we are trying to save a few dollars by switching to a capable open source database and running our failover systems with older 32-bit hardware rather than an identical new 64-bit machine. Thank you, Adam Talaat ------=_NextPart_000_002B_01C85F63.BEAABA40 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable
Hello,
 
Are the database files = created by=20 32-bit PostgreSQL identical to the files created by 64-bit PostgreSQL? = The=20 database files would reside on a SAN or DAS device external to the = servers. If=20 the server running a 64-bit installation of PostgreSQL were to fail, = would we be=20 able to bring a server running a 32-bit installation online and run it = against=20 the original database files?
 
We would absolutely not be = simultaneously running two instances of the database against the same = database=20 files. My background is with MS SQL Server, where a database can be = “reattached”=20 to a different instance of the software (although not between 32-bit and = 64-bit=20 versions). We would be running PostgreSQL on RHEL/CentOS=20 5.
 
I realize having a = separate server=20 is not a sufficient disaster recovery strategy. We do regular database = backups=20 to tape in the event of data corruption or catastrophic=20 failure.
 
Replication with Slony-I = is also an=20 option, however a complete hot or warm spare configuration would require = additional storage space. Is it possible to replicate from 64-bit to=20 32-bit?
 
If you haven’t = already guessed, we=20 are trying to save a few dollars by switching to a capable open source = database=20 and running our failover systems with older 32-bit hardware rather than = an=20 identical new 64-bit machine.
 
Thank = you,
Adam=20 Talaat
------=_NextPart_000_002B_01C85F63.BEAABA40--