From owner-freebsd-current Thu Nov 2 01:11:15 1995 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id BAA22786 for current-outgoing; Thu, 2 Nov 1995 01:11:15 -0800 Received: from utrhcs.cs.utwente.nl (utrhcs.cs.utwente.nl [130.89.10.247]) by freefall.freebsd.org (8.6.12/8.6.6) with SMTP id BAA22781 for ; Thu, 2 Nov 1995 01:11:10 -0800 Received: from myrtilos.cs.utwente.nl by utrhcs.cs.utwente.nl (5.x/csrelayMX-SVR4_1.1tmp/RB) id AA27655; Thu, 2 Nov 1995 10:10:49 +0100 Received: from curie.cs.utwente.nl by myrtilos.cs.utwente.nl (SMI-8.6/csrelay-Sol1.4/RB) id KAA21257; Thu, 2 Nov 1995 10:10:45 +0100 Received: from localhost by curie.cs.utwente.nl (SMI-8.6/SMI-SVR4) id KAA17471; Thu, 2 Nov 1995 10:10:45 +0100 To: michael butler Cc: current@freebsd.org Subject: Re: apparent tcp double-send ? In-Reply-To: Your message of "Thu, 02 Nov 1995 07:23:40 +1100." <199511012023.HAA19813@asstdc.scgt.oz.au> Date: Thu, 02 Nov 1995 10:10:44 +0100 Message-Id: <17470.815303444@curie.cs.utwente.nl> From: Andras Olah Sender: owner-current@freebsd.org Precedence: bulk On Thu, 02 Nov 1995 07:23:40 +1100, michael butler wrote: > I noticed this whilst doing a little snooping .. > > ppp1.asstdc.com.au.1091 > asstdc.scgt.oz.au.http: S 951910400:951910400(0) win 4096 (ttl 59, id 6771) > ppp1.asstdc.com.au.1091 > asstdc.scgt.oz.au.http: S 951910400:951910400(0) win 4096 (ttl 58, id 6771) > asstdc.scgt.oz.au.http > ppp1.asstdc.com.au.1091: S 4154539922:4154539922(0) ack 951910401 win 16384 (ttl 64, id 12805) > asstdc.scgt.oz.au.http > ppp1.asstdc.com.au.1091: S 4154539922:4154539922(0) ack 951910401 win 16384 (ttl 63, id 12805) [ ... ] > ppp1 is connected via another -current machine which uses the same address > for ethernet and ppp but no proxy arp. Strange MSS .. anyway, ppp1 is using > Trumpet 2.0b .. is this BSD, Trumpet or TcpDump ? Suggestions ? The duplicates are caused by some routing anomaly. Notice that the TTL of the duplicate is always one less than that of the original. As for the MSS, they're OK. Please notice that the behavior of FreeBSD changed by 2.0.5R, because the original BSD treatment of MSS was not correct. The MSS you send should depend only on the capabilities of your i/f, but not on the route. It's the responsibility of the other end, which recieves your MSS, to calculate its optimum outgoing segment size based on your MSS option and on the route's MTU. Andras