From owner-freebsd-questions Thu Jun 3 9:18:11 1999 Delivered-To: freebsd-questions@freebsd.org Received: from mailhub.scl.ameslab.gov (mailhub.scl.ameslab.gov [147.155.137.127]) by hub.freebsd.org (Postfix) with ESMTP id 2727614FF8 for ; Thu, 3 Jun 1999 09:18:05 -0700 (PDT) (envelope-from ghelmer@scl.ameslab.gov) Received: from demios.ether.scl.ameslab.gov ([147.155.137.54]) by mailhub.scl.ameslab.gov with esmtp (Exim 1.90 #1) id 10paCl-0002bg-00; Thu, 3 Jun 1999 11:18:55 -0500 Date: Thu, 3 Jun 1999 11:18:02 -0500 From: Guy Helmer To: Erik de Zeeuw Cc: freebsd-questions@freebsd.org Subject: Re: Bridging : "collision at xxx" exact meaning ? In-Reply-To: <37569FBB.86835FF@univ-lehavre.fr> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG On Thu, 3 Jun 1999, Erik de Zeeuw wrote: > I just compile a new kernel, using 3.2-RELEASE downloaded from > sunsite.org.uk, > with BRIDGE & DUMMYNET & IPFIREWALL enabled, default rule to accept. > > The machine has two Intel Pro 10/100 adapters, using driver fxp > (fxp0,fxp1). > > The bridge seems to work fine now. Everything's ok, except for the HUGE > number > of messages I have on console. > > These are : collision at xxx > > where "xxx" is a number, 99% of the time simply 0. > > I looked around at the code, and these are coming from bridge.c. After > studying the > code, it turns out to me that the bridge get a packet for wich it stored > an ethernet > adress, but the ethernet adress in the packet is different from the one > stored. This means that there was a "collision" in the hash table. Entries in the hash table are hashed by the two bytes of the Ethernet MAC address. A "collision" results when an entry in the hash table is re-used by a different MAC address. If you get this message very often, it means that the bridging code is working less efficiently that it otherwise would. When I encountered this problem, I increased HASH_SIZE in /sys/net/bridge.h to 16384 and it reduced the number of collisions. The hash table seems to be sparsely populated (there can't be 8192 machines on our department's net), so this seems to waste space :-( I'm not sure if this is the recommended solution, but it seemed to help. Guy Guy Helmer, Ph.D. Candidate, Iowa State University Dept. of Computer Science Research Assistant, Ames Laboratory --- ghelmer@scl.ameslab.gov Research Assistant, Dept. of Computer Science --- ghelmer@cs.iastate.edu http://www.cs.iastate.edu/~ghelmer To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message