| |
PA2AGA > HDDIG 28.02.00 01:32l 166 Lines 6512 Bytes #-9565 (0) @ EU
BID : HD_2000_57B
Read: GUEST
Subj: HamDigitalDigest 2000/57B
Path: DB0AAB<DB0FSG<DB0PV<OE2XOM<OE5XBL<OE6XAR<OE3XPR<OM0PBM<OM0PBB<SR9ZAA<
PI8JOP<PI8ZAA<PI8HWB<PI8HGL<PE1NMB<EA7URC<PE0MAR<PI8VNW
Sent: 000227/1704Z @:PI8VNW.#ZH2.NLD.EU #:54703 [HvHolland] FBB7.00g24
From: PA2AGA@PI8VNW.#ZH2.NLD.EU
To : HDDIG@EU
Received: from pa2aga by pi1hvh with SMTP
id AA31034 ; Sat, 26 Feb 00 17:55:21 UTC
Received: from pa2aga by pa2aga (NET/Mac 2.3.70/7.5.3) with SMTP
id AA00018135 ; Sat, 26 Feb 2000 15:33:43 MET
Date: Sat, 26 Feb 00 15:33:01 MET
Message-Id: <hd_2000_57B>
From: pa2aga
To: hd_broadcast@pa2aga
Subject: HamDigitalDigest 2000/57B
X-BBS-Msg-Type: B
> just now getting some decent little toys to take advantage of that
> sickly data transfer rate(the new Kenwood TH-d7a radio looks cool).
You forgot to mention that most of the Los Angeles area is still in the 1200
baud TNC world.
-------
Athough I can go back and look at the SCDCC minutes to see who you're talking
about, want to give me a hint? (private e-mail OK for this if you prefer.)
>.
------------------------------
Date: Fri, 25 Feb 2000 19:53:52 GMT
From: nomail@rob.knoware.nl (Rob Janssen)
Subject: AX.25 with TCP/IP routing?
Dale M. Skiba <ds0007@medtronic.COM> wrote:
>I haven't read the details on AX.25, but I have a guess as
>to what you are saying. Perhaps the root cause of the problem
>is that X.25 includes both data link and routing stuff.
First read the details... The AX.25 protocol has little or nothing to do
with the X.25 protocol!
>Although X.25 Packet Layer Protocol is interesting, history
>has shown that IP datagram routing is more flexible and robust.
>Thus, what's in wide spread use for the Internet is:
>TCP ISO level 4, "transport"
>IP ISO level 3, "network"
>ethernet ISO level 2, "data link"
>You seem to say they suggested redundant routing:
>TCP ISO level 4, "transport"
>IP ISO level 3, "network"
>AX.25 ISO level 3, "X.25 PLP network stuff"
>AX.25 ISO level 2, "X.25 HDLC/LAPB data link stuff"
There is no 'AX.25 ISO level 3, "X.25 PLP network stuff"', except
in the ROSE nodes which are not very widespread. Most nodes use NET/ROM
or FLEXNET, which are routing protocols that use a message structure
completely different from X.25 PLP.
>What they should have done was use only the data link part
>of X.25 and just use IP for the routing to keep things simple:
>TCP ISO level 4, "transport"
>IP ISO level 3, "network"
>AX.25 ISO level 2, "X.25 HDLC/LAPB"
You are missing one part: there should be a "service" that accepts
AX.25 connects (level 2, that is all there is) from users, interprets
their requests for further connects to other nodes and users (possibly
deriving this info from the "digipeater" fields) and tunnels the data
they send in a TCP session to the remote node, which sets up another
AX.25 connection to the end user.
This service allows end-users without interest in TCP/IP to connect to
other users or services that are AX.25-based via the network, while the
network itself is purely IP based. That means that other protocols can
be used at level 2 internal to the network, as long as they can serve as
a link level to transport IP datagrams. In some countries it may be
AX.25 (to avoid license hassles), in other countries it could be raw
HDLC or a better protocol that allows for FEC etc.
Rob
--
+----------------------------------+--------------------------------------+
| Rob Janssen pe1chl@amsat.org | WWW: http://www.knoware.nl/users/rob |
| AMPRnet: rob@pe1chl.ampr.org | AX.25 BBS: PE1CHL@PI8WNO.#UTR.NLD.EU |
+----------------------------------+--------------------------------------+
>.
------------------------------
Date: Fri, 25 Feb 2000 21:58:06 GMT
From: horseshoestew@my-deja.com
Subject: AX.25 with TCP/IP routing?
In article <896d33$f3c$1@gazette.corp.medtronic.com>,
ds0007@medtronic.COM (Dale M. Skiba) wrote:
> Rob Janssen (nomail@rob.knoware.nl) wrote:
>
> : On the land networks, there is no globally agreed standard protocol.
Not
> : even for routing AX.25, let alone TCP/IP.
> : (node software designers misjudge IP as a "level above AX.25" to be
> : routed "over the AX.25 routing mechanism in use", instead of the
other
> : way around)
>
> : Rob
>
> I haven't read the details on AX.25, but I have a guess as
> to what you are saying. Perhaps the root cause of the problem
> is that X.25 includes both data link and routing stuff.
>
> Although X.25 Packet Layer Protocol is interesting, history
> has shown that IP datagram routing is more flexible and robust.
> Thus, what's in wide spread use for the Internet is:
>
> TCP ISO level 4, "transport"
> IP ISO level 3, "network"
> ethernet ISO level 2, "data link"
>
> You seem to say they suggested redundant routing:
>
> TCP ISO level 4, "transport"
> IP ISO level 3, "network"
> AX.25 ISO level 3, "X.25 PLP network stuff"
> AX.25 ISO level 2, "X.25 HDLC/LAPB data link stuff"
>
> What they should have done was use only the data link part
> of X.25 and just use IP for the routing to keep things simple:
>
> TCP ISO level 4, "transport"
> IP ISO level 3, "network"
> AX.25 ISO level 2, "X.25 HDLC/LAPB"
Man, you left out NETROM nodes. I don't know about today's networks(got
out of it 5 years ago), but it used to be that the "control freaks"
that were in charge of key networking sites loved their silly NETROM
nodes(that gave them little dictator status), and weren't gonna give
them up for nothing.
Then there were the IP coordinator Hitlers and Mussolinis. They used to
ignorantly, greedily and stupidly hoard their huge banks of unused IP
address allocations, while amateurs were being threatened with having to
give up there ample IP address allocation - because they weren't using
them! One time, I was trying to set up subnets for a "serious" cross
frequency network in Riverside County - and my coordinator just didn't
get it, even when I totally spelled it out in writing, and he wouldn't
give me any 8-bit subnet address allocations - so I just said "screw
'em" and picked my own. He went running to Phil Karn, or the National
IP address coordinator, or somebody or other, who helped him understand
I was doing the right thing - so he finally(weeks later) relented to my
"choice" of subnet addresses. It was way up in his 16-bit address
allocation block, and I guess he finally saw that it posed "no threat"
to him and his "control" efforts of the uneducated masses.
You've got to understand that this whole amateur packet radio thing is
by definition experimental and is COMPLETELY OUT OF CONTROL. Everyone
goes with their own thing, kinda just playing around. Don't EVEN think
To be continued in digest: hd_2000_57C
Read previous mail | Read next mail
| |
