From mboxrd@z Thu Jan 1 00:00:00 1970 From: jamal Subject: Re: Routing 600+ vlan's via linux problems (looks like arp problems) Date: Thu, 03 May 2007 20:01:38 -0400 Message-ID: <1178236898.4068.47.camel@localhost> References: <20070503205346.GD18539@uranus.ravnborg.org> <20070503143832.17a08fbb@freekitty> Reply-To: hadi@cyberus.ca Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Cc: Sam Ravnborg , =?ISO-8859-1?Q?=D8yvind_V=E5gen_J=E6gtnes?= , netdev To: Stephen Hemminger Return-path: Received: from wx-out-0506.google.com ([66.249.82.225]:44128 "EHLO wx-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1767473AbXEDABo (ORCPT ); Thu, 3 May 2007 20:01:44 -0400 Received: by wx-out-0506.google.com with SMTP id h31so598713wxd for ; Thu, 03 May 2007 17:01:43 -0700 (PDT) In-Reply-To: <20070503143832.17a08fbb@freekitty> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Thu, 2007-03-05 at 14:38 -0700, Stephen Hemminger wrote: > What kernel version? Are you on a recent 2.6 kernel or stuck on some > old "vendor stable" 2.4 kernel? VendorStable(tm)?;-> Sounds to me like an ARP gc challenge to me. If it is, it would help incrementing the values in /proc/sys/net/ipv4/neigh/default/gc_thresh* Try multiplying by some factor like 5 to see what happens i.e --- hadi@lilsol:~$ cat /proc/sys/net/ipv4/neigh/default/gc_thresh1 128 --- Make that 640 and repeat multiplying by 5 all the other thresholds cheers, jamal PS:- As usual i took lkml off the list