X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.2]) by svr1.postgresql.org (Postfix) with ESMTP id 35F4DD1BAF5 for ; Tue, 20 Apr 2004 23:14:54 -0300 (ADT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (neptune.hub.org [200.46.204.2]) (amavisd-new, port 10024) with ESMTP id 25195-10 for ; Tue, 20 Apr 2004 23:14:55 -0300 (ADT) Received: from hotmail.com (law10-oe51.law10.hotmail.com [64.4.14.40]) by svr1.postgresql.org (Postfix) with ESMTP id AF845D1B974 for ; Tue, 20 Apr 2004 23:14:51 -0300 (ADT) Received: from mail pickup service by hotmail.com with Microsoft SMTPSVC; Tue, 20 Apr 2004 19:14:51 -0700 Received: from 67.81.102.201 by law10-oe51.law10.hotmail.com with DAV; Wed, 21 Apr 2004 02:14:50 +0000 X-Originating-IP: [67.81.102.201] X-Originating-Email: [awerman2@hotmail.com] X-Sender: awerman2@hotmail.com From: "Aaron Werman" To: "Joshua D. Drake" , "Anjan Dave" Cc: References: <4BAFBB6B9CC46F41B2AD7D9F4BBAF78509814D@vt-pe2550-001.vantage.vantage.com> <4085BFF0.4050604@commandprompt.com> Subject: Re: Moving postgres to FC disks Date: Tue, 20 Apr 2004 22:14:51 -0400 MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0064_01C42724.E65B1010" X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1409 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Message-ID: X-OriginalArrivalTime: 21 Apr 2004 02:14:51.0302 (UTC) FILETIME=[6D96F460:01C42746] X-Virus-Scanned: by amavisd-new at postgresql.org X-Spam-Status: No, hits=1.2 tagged_above=0.0 required=5.0 tests=HTML_40_50, HTML_MESSAGE, HTML_TITLE_EMPTY, RCVD_IN_NJABL, RCVD_IN_SORBS X-Spam-Level: * X-Archive-Number: 200404/326 X-Sequence-Number: 6626 This is a multi-part message in MIME format. ------=_NextPart_000_0064_01C42724.E65B1010 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable I agree on not linking and adding non-SAN disk dependancy to your DB. I'm t= rying to understand your FS reasoning. I have never seen XFS run faster tha= n ReiserFS in any situation (or for that matter beat any FS in performance = except JFS). XFS has some nifty very large file features, but we're talking= about 30G and all modern FSs support >2G files.=20 My tendancy would be to stay on ext3, since it is the default RH FS. I woul= d review site preference and the SAN recommended FS and see if they add any= compelling points. /Aaron ----- Original Message -----=20 From: Joshua D. Drake=20 To: Anjan Dave=20 Cc: pgsql-performance@postgresql.org=20 Sent: Tuesday, April 20, 2004 8:27 PM Subject: Re: [PERFORM] Moving postgres to FC disks -With the db size being as big as, say, 30+GB, how do I move it on the = new logical drive? (stop postgresql, and simply move it over somehow and ma= ke a link?) I would stop the database, move the data directory to the new volume usin= g rsync then start up postgresql pointed at the new data directory. Providing everything is working correctly you can then remove the old dat= a directory. -Currently, the internal RAID volume is ext3 filesystem. Any recommenda= tions for the filesystem on the new FC volume? Rieserfs? XFS DBs are 7.4.1(RH9), and 7.2.3 (RH8). Appreciate any pointers. Thanks, Anjan --=20 Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC Postgresql support, programming shared hosting and dedicated hosting. +1-503-667-4564 - jd@commandprompt.com - http://www.commandprompt.com PostgreSQL Replicator -- production quality replication for PostgreSQL ------=_NextPart_000_0064_01C42724.E65B1010 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
I agree on not linking and adding non-SAN = disk=20 dependancy to your DB. I'm trying to understand your FS reasoning. <= FONT=20 face=3DArial size=3D2>I have never seen XFS run faster than ReiserFS in any= =20 situation (or for that matter beat any FS in performance except JFS). XFS h= as=20 some nifty very large file features, but we're talking about 30G and all mo= dern=20 FSs support >2G files.
=  
My tendancy would be to stay on ext3, sinc= e it is=20 the default RH FS. I would review site preference and the SAN recommended F= S and=20 see if they add any compelling points.
 
/Aaron
----- Original Message -----
Fro= m:=20 Josh= ua D.=20 Drake
Cc: pgsql-performance@postgr= esql.org=20
Sent: Tuesday, April 20, 2004 8:27= =20 PM
Subject: Re: [PERFORM] Moving post= gres to=20 FC disks


-With t= he db=20 size being as big as, say, 30+GB, how do I move it on the new logical d= rive?=20 (stop postgresql, and simply move it over somehow and make a=20 link?)

I would stop the database, m= ove the=20 data directory to the new volume using rsync then start up postgresql poi= nted=20 at the new data directory.
Providing everything is working correctly y= ou=20 can then remove the old data directory.

-Curren= tly,=20 the internal RAID volume is ext3 filesystem. Any recommendations for th= e=20 filesystem on the new FC volume? Rieserfs?

XFS

DBs are= =20 7.4.1(RH9), and 7.2.3 (RH8).

Appreci= ate any=20 pointers.

Thanks,=
Anjan



--=20
Command Prompt, Inc., home of Mammoth PostgreSQL - S/ODBC and S/JDBC
Postgresql support, programming shared hosting and dedicated hosting.
+1-503-667-4564 - jd@commandprompt.com - http://www.commandprompt.com
PostgreSQL Replicator -- production quality replication for PostgreSQL

------=_NextPart_000_0064_01C42724.E65B1010--