From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Hemminger Subject: Re: 2.6.23.1-smp kernel panic (network-related) Date: Wed, 7 Nov 2007 10:47:11 -0800 Message-ID: <20071107104711.03d1371c@freepuppy.rosehill> References: <20071107095120.237366d2@freepuppy.rosehill> <20071107193843.1f3fa2f4@catlap> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org To: Marek Kierdelewicz Return-path: Received: from smtp2.linux-foundation.org ([207.189.120.14]:38699 "EHLO smtp2.linux-foundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759291AbXKGSsc (ORCPT ); Wed, 7 Nov 2007 13:48:32 -0500 In-Reply-To: <20071107193843.1f3fa2f4@catlap> Sender: netdev-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Wed, 7 Nov 2007 19:38:43 +0100 Marek Kierdelewicz wrote: > >What is the test input that causes the crash?? > > Test box is treated with mirrored traffic that is routed by production > linux router (with non-smp kernel). It's usual traffic generated by > broadband clients. Some of the characteristics: > > bandwidth used: ~40/40 Mbit (up/down) > pps: ~15k > > number of clients: ~550 > > dump of packet sizes: > Packet size | Count > 1 to 75: 186501 > 76 to 150: 14145 > 151 to 225: 3285 > 226 to 300: 2088 > 301 to 375: 3632 > 376 to 450: 2097 > 451 to 525: 1513 > 526 to 600: 3069 > 601 to 675: 20081 > 676 to 750: 1294 > 751 to 825: 1189 > 826 to 900: 885 > 901 to 975: 2207 > 976 to 1050: 1333 > 1051 to 1125: 1192 > 1201 to 1275: 3036 > 1276 to 1350: 3709 > 1351 to 1425: 3453 > 1426 to 1500+: 185318 > > protocol breakdown: > most of the traffic is IPv4, some UDP and a little bit of ICMP > > Don't know if it's important, but box is connected to switch by a > 802.1q trunk. Each of vlan interfaces has egress shaping on it + > dedicated ifb device attached to ingress qdisc and ingress shaping on > ifb device. > > Is any additional information needed? Could you get 'ethtool -S' output after it hangs. I think there is a problem with fifo overflow. I can cause board to latch if enough traffic arrives on both ports. -- Stephen Hemminger