From insl2.etec.uni-karlsruhe.de!erb Thu Aug 31 20:04:34 1995
Return-Path: <erb@insl2.etec.uni-karlsruhe.de>
Received: from insl2.etec.uni-karlsruhe.de by iiit.swan.ac.uk  with smtp
	(Linux Smail3.1.28.1 #8) id m0soEua-00013hC; Thu, 31 Aug 95 20:04 BST
Received: (from erb@localhost) by insl2.etec.uni-karlsruhe.de id UAA00357
  (8.6.12/IDA-1.6 for iialan@iifeak.swan.ac.uk); Thu, 31 Aug 1995 20:37:00 +0200
Message-ID: <199508311837.UAA00357@insl2.etec.uni-karlsruhe.de>
Subject: Re: route with irtt support?
To: iialan@iifeak.swan.ac.uk (Alan Cox)
Date: Thu, 31 Aug 1995 20:36:59 +0200 (MET DST)
In-Reply-To: <m0soDdG-00013gC@iiit.swan.ac.uk> from "Alan Cox" at Aug 31, 95 06:42:29 pm
From: erb@inss1.etec.uni-karlsruhe.de (Olaf Erb)
Reply-To: erb@inss1.etec.uni-karlsruhe.de
X-Mailer: ELM [version 2.4 PL24 ME7a]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 3299      
Status: RO

Alan Cox writes:
> 
> > Very nice.. Alan also mentioned it somewhere, but I can't find a route
> > that supports this. Can someone give me a pointer?
> 
> net-tools-1.3.6BETA3, but remember its a beta.

Can't find it :(

% zcat net-tools-1.3.6-BETA3.tar.gz | grep -i irtt

Alan devra ajouter l'option 'irtt' (initial round trip time, c.a.d le
Alan needs to add the 'irtt' option for an initial round trip time for slow

The 'route' there don't recognise the irtt keyword..

Would be very nice for our dog-slow packet links here..

> > Btw, am I the only one who noticed that raw sockets are broken in the
> > newer 1.3er kernels? (frames are trunceted, using WAMPES and axip/ipip)
> 
> Which kernels. I'm not seeing it. I've done some more changes to raw
> sockets and may have cured it by mistake tho in adding IP_HDRINCL.

I tried it with 1.3.6(?) and later.. Now with 1.3.21, I'm still getting
this, rerouted the frame from ethernet to the KISS interface for
better tracing (through a mode VC connection):

trace on the axip interface:

Thu Aug 31 20:13:09 1995 - axip sent:
AX25: DC1IK-10->DK5SG SABM(P)

This becomes, on KISS:

dc1ik-1.ampr.org> c dk5sg
Conn pend

Thu Aug 31 20:10:29 1995 - ttyS2 sent:
KISS: crc-rmnc Data
AX25: DC1IK-10->DL6IU-2 v DC1IK-6 I NR=5 NS=1 pid=IP
IP: len 37 dc1ik.ampr.org->dk5sg-u.ampr.org ihl 20 ttl 63 prot 0
0000  ..j..@...b..@u?.Z


Or with Hex-Trace, the same (SABM):

Thu Aug 31 20:12:08 1995 - ttyS2 sent:
KISS: crc-rmnc Data
AX25: DC1IK-10->DL6IU-2 v DC1IK-6 I NR=3 NS=4 pid=IP
IP: len 37 dc1ik.ampr.org->dk5sg-u.ampr.org ihl 20 ttl 63 prot 0
0000  20 88 98 6c 92 aa 40 e4 88 86 62 92 96 40 74 88   ..l..@...b..@t.
0010  86 62 92 96 40 6d 68 cc 45 00 00 25 00 42 00 00  .b..@mh.E..%.B..
0020  3f 00 c9 2a 2c 82 29 0a 2c 82 30 5f 88 96 6a a6  ?..*,.).,.0_..j.
0030  8e 40 e0 88 86 62 92 96 40 75 3f 7f 5a 2a 04     .@...b..@u?.Z*.

Receiving axip frames still works.
Sending frames is ok under 1.2.13. I'll reboot now and include it here:

Thu Aug 31 20:30:06 1995 - ttyS2 sent:
KISS: crc-rmnc Data
AX25: DC1IK-10->DL6IU-2 v DC1IK-6 I NR=7 NS=1 pid=IP
IP: len 37 dc1ik.ampr.org->dk5sg-u.ampr.org ihl 20 ttl 63 prot AX25
AX25: DC1IK-10->DK5SG SABM(P)
0000  .Z

HEX:
Thu Aug 31 20:30:58 1995 - ttyS2 sent:
KISS: crc-rmnc Data
AX25: DC1IK-10->DL6IU-2 v DC1IK-6 I NR=5 NS=2 pid=IP
IP: len 37 dc1ik.ampr.org->dk5sg-u.ampr.org ihl 20 ttl 63 prot AX25
AX25: DC1IK-10->DK5SG SABM(P)
0000  20 88 98 6c 92 aa 40 e4 88 86 62 92 96 40 74 88   ..l..@...b..@t.
0010  86 62 92 96 40 6d a4 cc 45 00 00 25 00 3c 00 00  .b..@m..E..%.<..
0020  3f 5d c8 d3 2c 82 29 0a 2c 82 30 5f 88 96 6a a6  ?]..,.).,.0_..j.
0030  8e 40 e0 88 86 62 92 96 40 75 3f 7f 5a 36 cf     .@...b..@u?.Z6.

Hmmm..
The proto 93 (0021:0x5d) is changed to 00. Framelength is the same, so 
no truncation as I stated before, but something changed the data.
(maybe bytes 21 to 23).

I'm sorry that I can't find the problem myself, but with a quick
overlook I don't understand the changes in the raw socket code, I'm
no kernel hacker :(

73,
Olaf
-- 
---------------------------------------------------------------
!   erb@inss1.etec.uni-karlsruhe.de  dc1ik@db0sao.ampr.org    !
! <A HREF="http://inss1.etec.uni-karlsruhe.de/~erb">click</A> !
---------------------------------------------------------------

