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 8815A53550 for ; Thu, 31 Mar 2005 15:00:50 +0100 (BST) 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 39470-04 for ; Thu, 31 Mar 2005 14:00:44 +0000 (GMT) Received: from vms044pub.verizon.net (vms044pub.verizon.net [206.46.252.44]) by svr1.postgresql.org (Postfix) with ESMTP id F242653548 for ; Thu, 31 Mar 2005 15:00:43 +0100 (BST) Received: from jfradkin ([66.15.127.235]) by vms044.mailsrvcs.net (Sun Java System Messaging Server 6.2 HotFix 0.04 (built Dec 24 2004)) with ESMTPA id <0IE7008ALYXAC520@vms044.mailsrvcs.net> for pgsql-admin@postgresql.org; Thu, 31 Mar 2005 08:00:47 -0600 (CST) Date: Thu, 31 Mar 2005 09:00:59 -0500 From: "Joel Fradkin" Subject: Re: connectivity to database been an issue for me. In-reply-to: <16971.45257.133152.867195@tanto-xipolis.de> To: "'Morus Walter'" , Message-id: <001401c535fa$1151b230$797ba8c0@jfradkin> MIME-version: 1.0 X-MIMEOLE: Produced By Microsoft MimeOLE V6.00.2900.2180 X-Mailer: Microsoft Outlook, Build 10.0.6626 Content-type: text/plain; charset=us-ascii Content-transfer-encoding: quoted-printable Importance: Normal X-Priority: 3 (Normal) X-MSMail-priority: Normal X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=1.477 tagged_above=0 required=5 tests=AWL X-Spam-Level: * X-Archive-Number: 200503/504 X-Sequence-Number: 17002 Well the issue seems to occur for any traffic not limited to any one machine. Pinging Yahoo for instance the ping is timing out when the line gets = full load. I spoke to Verizon and they admitted they sold more bandwidth then the = hard wires in the area support. I am going to see if we can slow down the traffic at the fire wall, but = also plan on getting a 2mb breakfree wireless account as it is just 100.00 a month and much better connectivity. My real question was if it would be better to get a T1 to the data = center or if going over the internet was ok if I get the expanded bandwidth; based = on real world experience doing a similar thing. I hate to recommend an expensive solution just to end up with the same issues. I tried loading = the line with various types of communication to different machines, so it is = not a postgres issue at all (sorry for voicing the question in a pg forum, = but I figured you guys deal with a bit of the networking as well). Unfortunatley the T1 is a bit expensive (T1>500.00 a month). Joel Fradkin =20 -----Original Message----- From: Morus Walter [mailto:morus.walter@tanto.de]=20 Sent: Thursday, March 31, 2005 3:12 AM To: pgsql-admin@postgresql.org; Joel Fradkin Subject: Re: [ADMIN] connectivity to database been an issue for me. Joel Fradkin writes: > Hi, >=20 > =20 >=20 > We are in development stage of implementing postgres. >=20 > Our data center is in Tampa and we are in Bardenton and I am using DSL = at my > office (384 up, 1.5 down). >=20 > Data center is 2 meg ea way. >=20 > When I use pgadmin to connect it is ok, but as soon as I do any querry that > has much data I get disconeected. >=20 > I was able to test pinging an outside web page (yahoo.com) and could = see as > soon as I start the transimission from a query in pgadmin the pinging = gets > disrupted (also I lose the connection to pgadmin). >=20 > =20 >=20 > I am thinking of T1 to data center (but is like 600 a month) or even upping > the office to breakfreee which is pretty cheap and is 2 mb ea way. >=20 > Before I spend $ I thought I would ask if there are any tools less = subject > to disconnect from temporary connection loss? >=20 > If any one else experienced the same type of issues and what they did = to > resolve the issue. >=20 As others already pointed out, this is likely to be a networking problem that has nothing to do with postgresql. What comes in my mind regarding DSL is that a common error is to use a too large MTU. Since there is some repackaging of network packets in the ppp over ethernet protocol, one needs to make sure that the raw = packages are small enough to allow for the overhead. Otherwise the effect is, that small packages (e.g. your ping) pass, = while large packages hang the connection. Don't ask me for details of the configuration, I do that ~ once every = two years (on linux). Morus