From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 4 Jul 2002 08:35:57 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 4 Jul 2002 08:35:56 -0400 Received: from aboukir-101-1-23-willy.adsl.nerim.net ([62.212.114.60]:10003 "EHLO www.home.local") by vger.kernel.org with ESMTP id ; Thu, 4 Jul 2002 08:35:55 -0400 Date: Thu, 4 Jul 2002 14:38:22 +0200 From: Willy Tarreau To: gphat@cafes.net Cc: linux-kernel@vger.kernel.org Subject: Re: Large numbers of TCP resets Message-ID: <20020704123822.GA26203@alpha.home.local> References: <20020703201553.DE9FD68CB5EA@mail.cafes.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20020703201553.DE9FD68CB5EA@mail.cafes.net> User-Agent: Mutt/1.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Jul 03, 2002 at 08:15:53PM +0000, gphat@cafes.net wrote: > Recently, the web-app at the company I work for started having problems load > balancing. This was traced back to a large number of tcp-resets being sent > from the web servers to the clients. aren't your load balancers Alteon ACE Directors ? If this is the case, I suspect you use "fastage 0", which ends sessions prematurely. This particularly happens in case of direct access mode (DAM) because the switch needs to remap source ports, and when the session expires, it simply routes packets from server to client without DNATing them. The client then receives ACKs and/or FINs for closed ports, or for open ports, but with bad sequence numbers, and then sends RESETS, which the server doesn't understand. I observed real RST storms during tens of minutes because of this. They disappeared when I set "fastage" to something higher than 4 (=keep the session at least 16 seconds, even in final states). I think this is not specific to Linux 2.4 since I observed a similar behaviour with Solaris 8. Cheers, Willy