From owner-freebsd-questions Sun Apr 13 19:38:21 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA10174 for questions-outgoing; Sun, 13 Apr 1997 19:38:21 -0700 (PDT) Received: from icicle.winternet.com (adm@icicle.winternet.com [198.174.169.5]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA10164 for ; Sun, 13 Apr 1997 19:38:17 -0700 (PDT) Received: (from adm@localhost) by icicle.winternet.com (8.7.5/8.7.5) id VAA09194 for ; Sun, 13 Apr 1997 21:38:08 -0500 (CDT) Posted-Date: Sun, 13 Apr 1997 21:38:08 -0500 (CDT) Received: from fools.ecp.net(204.246.64.101) by icicle.winternet.com via smap (V2.0beta) id xma009178; Sun, 13 Apr 97 21:37:58 -0500 Received: from localhost (moke@localhost) by fools.ecp.net (8.8.5/8.8.4) with SMTP id VAA06536 for ; Sun, 13 Apr 1997 21:37:49 -0500 (CDT) Date: Sun, 13 Apr 1997 21:37:49 -0500 (CDT) From: Jimbo Bahooli To: freebsd-questions@freebsd.org Subject: update: ipnat + natd Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-questions@freebsd.org X-Loop: FreeBSD.org Precedence: bulk About a week ago I posted a message about dismal perfomance on translated hosts with ipnat + natd. (20bytes/sec) This was on a 2.2-STABLE system using kernel ppp with an mtu/mru size of 296. To find the solution I tried a few things, one of them i bumped the mru/mtu up to 552. This instantly brought translated througput to 1.9-2.3k/sec which is acceptable on a 28.8 connect. I then brought down the ethernet mtu to 552, this increased the reaction time somewhat, and possibly raised througput alittle more. User ppp might also suffer from this, my personal recommendation now is to keep the mtu/mru sizes the same between the translated interfaces.