From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: iproute uses too small of a receive buffer Date: Wed, 28 Oct 2009 21:38:46 +0100 Message-ID: <4AE8ABD6.2010203@gmail.com> References: <4AE77F64.3090302@candelatech.com> <20091027162434.6dc31b2d@nehalam> <4AE7F859.7020105@gmail.com> <4AE895E8.60308@trash.net> <4AE89927.9090405@candelatech.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Patrick McHardy , Stephen Hemminger , NetDev To: Ben Greear Return-path: Received: from gw1.cosmosbay.com ([212.99.114.194]:56320 "EHLO gw1.cosmosbay.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755471AbZJ1Uiq (ORCPT ); Wed, 28 Oct 2009 16:38:46 -0400 In-Reply-To: <4AE89927.9090405@candelatech.com> Sender: netdev-owner@vger.kernel.org List-ID: Ben Greear a =E9crit : > Second: Why bail on ENOBUFS at all? I don't see how it helps the us= er > since they will probably just have to start it again, and will miss m= ore > messages than keeping going would have. >=20 > And, even 1MB may not be enough for some scenarios. So, probably bes= t to > let users over-ride the initial setting on cmd-line. If not, then us= e > a large value to start with. >=20 In this case, just dont call setsockopt() at all in "ip" and let system= use the standard/default value (/proc/sys/net/core/rmem_default) that an admin= can change if he wants to handle one million devices :)