Received: from localhost (unknown [200.46.204.184]) by postgresql.org (Postfix) with ESMTP id 54FE42E0065 for ; Thu, 20 Mar 2008 17:46:38 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.184]) (amavisd-maia, port 10024) with ESMTP id 24938-06 for ; Thu, 20 Mar 2008 17:45:59 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from spider01.iglass.net (ns1.iglass.net [66.192.95.35]) by postgresql.org (Postfix) with ESMTP id D08C62E0068 for ; Thu, 20 Mar 2008 17:45:20 -0300 (ADT) Received: from iglassnoc3 (ig128.iglass.net [192.168.177.128] (may be forged)) by spider01.iglass.net (8.13.1/8.13.1) with ESMTP id m2KKjDh9026415; Thu, 20 Mar 2008 16:45:13 -0400 From: "Woody Woodring" To: "'Chris Hoover'" , "'PGSQL Performance'" References: <1d219a6f0803201232g54924fb3ka206c9b03e4664c7@mail.gmail.com> Subject: Re: PostgreSQL NetApp and NFS Date: Thu, 20 Mar 2008 16:45:14 -0400 Message-ID: <048001c88acb$4c2b27f0$80b1a8c0@istructure.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0481_01C88AA9.C51987F0" X-Mailer: Microsoft Office Outlook 11 In-Reply-To: <1d219a6f0803201232g54924fb3ka206c9b03e4664c7@mail.gmail.com> X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.3198 Thread-Index: AciKw7iTkqJJonM1T0G3CtQ7t3GF1gAAfGNQ X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200803/413 X-Sequence-Number: 29367 This is a multi-part message in MIME format. ------=_NextPart_000_0481_01C88AA9.C51987F0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit > I need to know if anyone out there is/has run their PostgreSQL on NetApp arrays via NFS. My particular situation is RH Linux 4 servers running Postgresql 8.1. I need > to provide our Operations manager with specific reasons why we should not run PostgreSQL over NetApp NFS. Otherwise, they will go forward with this. > If you have any real life good or bad stories, I'd love to hear it. Given the NetApp arrays supposedly being very good NFS platforms, overall, is this a recommended way > to run PostgreSQL, or is it recommended to not run this way. We have been running Postgres over NFS to a NetApp since 7.1 and we have nothing but good things to say. We have 75 databases in 3 clusters all connected to one netapp. We don't store a huge amount of data, currently ~43Gig, but it is constantly updated. We keep the pgsql/data directory on the netapp. If one of our db servers ever have a problem, we can just swap out the box, mount the drive and restart postgres. We like our support we get from them, the only issue we ever have is having a drives fail which they get replacements to us promptly. Our NetApp has an uptime currently over 2 years. By the way, I though NetApp boxes came with an iSCSI license. NetApp downplayed the iSCSI with us because they said you cannot share drives between servers, but for postgres you don't want that anyway. It could have also been that the NetApp is better tuned for NFS throughput and they want to steer the user toward that. If you want more specifics, feel free to ask. Woody iGLASS Networks ------=_NextPart_000_0481_01C88AA9.C51987F0 Content-Type: text/html; charset="US-ASCII" Content-Transfer-Encoding: quoted-printable

> I need to know if anyone out = there is/has=20 run their PostgreSQL on NetApp arrays via NFS. My particular = situation is=20 RH Linux 4 servers running Postgresql 8.1. I need

> to provide our = Operations manager=20 with specific reasons why we should not run PostgreSQL over NetApp = NFS. =20 Otherwise, they will go forward with this.

> If=20 you have any real life good or bad stories, I'd love to hear it. = Given the=20 NetApp arrays supposedly being very good NFS platforms, overall, is this = a=20 recommended way

> to run PostgreSQL, or = is it=20 recommended to not run this way.

We have been running Postgres over NFS to = a NetApp=20 since 7.1 and we = have nothing=20 but good things to say. We have 75 databases in 3 clusters all = connected=20 to one netapp. We don't store a huge amount of data, currently = ~43Gig, but=20 it is constantly updated.

We keep the pgsql/data directory on the = netapp. If=20 one of our db servers ever have a problem, we can just swap out the box, = mount=20 the drive and restart postgres.

We like our support we get from them, the only = issue we=20 ever have is having a drives fail which they get replacements to us=20 promptly. Our NetApp has an uptime currently over 2=20 years.

By the way, I though NetApp boxes came = with an iSCSI=20 license. NetApp downplayed the iSCSI with us because they = said you=20 cannot share drives between servers, but for postgres you don't = want that=20 anyway. = It could have=20 also been that the NetApp is better tuned for NFS throughput and they = want to=20 steer the user toward that.

If you want more specifics, feel free to=20 ask.

Woody

iGLASS Networks

------=_NextPart_000_0481_01C88AA9.C51987F0--