Received: from localhost (wm.hub.org [200.46.204.128]) by postgresql.org (Postfix) with ESMTP id BD7DD9FB24B for ; Mon, 2 Oct 2006 12:50:10 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.128]) (amavisd-new, port 10024) with ESMTP id 94814-05 for ; Mon, 2 Oct 2006 15:49:57 +0000 (UTC) X-Greylist: domain auto-whitelisted by SQLgrey- Received: from ug-out-1314.google.com (ug-out-1314.google.com [66.249.92.169]) by postgresql.org (Postfix) with ESMTP id 45ECD9FB20D for ; Mon, 2 Oct 2006 12:49:56 -0300 (ADT) Received: by ug-out-1314.google.com with SMTP id 30so500006ugc for ; Mon, 02 Oct 2006 08:49:55 -0700 (PDT) DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws; s=beta; d=gmail.com; h=received:message-id:date:from:to:subject:mime-version:content-type:content-transfer-encoding:content-disposition; b=pIOsQRBnKI8sOgn2GXqVRHgvPhCP1V9NDMtqKA51/dm4/7y2ygmmn5jKZ19I2MQ54eWa8ZM2SYVvv5MoyRWnMKTWUzArznpPMCJCaktq36KgLlt+NeVOay2RD8TQjln/epGHXevi/CUfYbiTC8BW/w/HFqKl5C/P2V0oTac0XL0= Received: by 10.67.89.5 with SMTP id r5mr2762653ugl; Mon, 02 Oct 2006 08:49:55 -0700 (PDT) Received: by 10.66.223.9 with HTTP; Mon, 2 Oct 2006 08:49:55 -0700 (PDT) Message-ID: Date: Mon, 2 Oct 2006 12:49:55 -0300 From: "Marcelo Costa" To: pgsql-performance@postgresql.org Subject: How much memory in 32 bits Architecture to Shared Buffers is Possible MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200610/4 X-Sequence-Number: 21133 Hi, to all! Recently i try increasing the memory values of shared buffers on one IBM xseries 255 (Quad XEON 2.8, 8 GB RAM, 2 disk SCSI 36 GB(Raid 1), 1 Storage. I try change these shared memory values to use 25% of memory ram (2048 MB) and effective_cache_size to 50% (4096 MB) of memory. All this settings to 220 Max Connections. Where I start up the cluster very messages of configurations errors on shared_memmory and SHMMAX look up. I try change the values of shared_memory, max_connections and effective_cache_size and large the size of SHMALL and SHMMAX to use 4294967296 (4096 MB) but the cluster don't start. Only with 15% of value on shared memory i can start up this cluster. In my tests the maximum value who i can put is 1.9 GB, more of this the cluster don't start. Can anybody help me and explicate if exist one limit to memory on 32 bits Architecture. Anybody was experience with tuning servers with this configurations and increasing ? thanks to all. -- Marcelo Costa