| |
ZL3AI > APRDIG 28.07.03 11:55l 242 Lines 9215 Bytes #999 (0) @ WW
BID : 1690-ZL3AI
Read: GUEST
Subj: TAPR Digest, Jul 25, 5/6
Path: DB0FHN<DB0RGB<OK0PPL<DB0RES<ON0AR<7M3TJZ<JE7YGF<CT2GWY<W4JAX<KD4GBA<
ZL2BAU<ZL3VML
Sent: 030727/1436Z @:ZL3VML.#80.NZL.OC #:5260 [Chch-NZ] FBB7.00i $:1690-ZL3AI
From: ZL3AI@ZL3VML.#80.NZL.OC
To : APRDIG@WW
Subject: Re: Speaking of Bob's discussion of RF usage....
From: Bob Bruninga <bruninga@usna.edu>
Date: Fri, 25 Jul 2003 14:43:19 -0400 (EDT)
X-Message-Number: 27
On Fri, 25 Jul 2003, Dick Stanich wrote:
>This is the type of awareness that pays off.... As you see, our local
>RF traffic is in the 'green', mostly. http://traffic.kb7zva.com
I applaud this effort, but somehow we need each such page to include the
following warning or words to this effect, so that users do not get the
wrong impression:
WARNING:
"Categorizing network behavior by the statistic of the number of
successful packets per hour can be completely missleading. A GREEN
indication of only 700 packets per hour throughput CAN mean a
lightly loaded network OR it can also indicate a totally swamped, totally
unreliable network suffering so many collisions that only 700 get
through."
The only reliable statistic is measuring the number of packets received
versus the number transmitted. Fortunately, this is now easy to do
by looking for PHG packets that are in the "PHGr" format. The "R"
indicates how often that packet is transmitted per hour. If the number
received per hour is less than the number transmitted, then that ratio is
the reliability of the network..
Good luck!
Bob
----------------------------------------------------------------------
Subject: RE: Laptop alternative
From: "Martin Maxwell" <maxwell679@cox.net>
Date: Fri, 25 Jul 2003 11:50:27 -0700
X-Message-Number: 28
I checked the website they are currently out of production as they are
upgrading to a faster processor to support Win XP..
Martin
N7HAB
Phoenix AZ
----------------------------------------------------------------------
Subject: Re: Speaking of Bob's discussion of RF usage....
From: "Dick Stanich" <dick@kb7zva.com>
Date: Fri, 25 Jul 2003 14:3:17
X-Message-Number: 29
>>This is the type of awareness that pays off.... As you see, our local
>>RF traffic is in the 'green', mostly. http://traffic.kb7zva.com
>
>I applaud this effort, but somehow we need each such page to include the
>following warning or words to this effect, so that users do not get the
>wrong impression:
>
>WARNING:
>
>"Categorizing network behavior by the statistic of the number of
>successful packets per hour can be completely missleading. A GREEN
>indication of only 700 packets per hour throughput CAN mean a
>lightly loaded network OR it can also indicate a totally swamped, totally
>unreliable network suffering so many collisions that only 700 get
>through."
I think someone else will explain it better Bob. It measures all 'traffic'
heard, doesn't determine whether packets are valid or not. Everytime the
TNC detects a packet signal, corrupt or collision, it is counted. If the
TNC lights up, it is traffic.
Dick, KB7ZVA
APRSWest
----------------------------------------------------------------------
Subject: Re: APRS Network Protection.
From: "Scott Miller" <scott@opentrac.org>
Date: Fri, 25 Jul 2003 12:16:39 -0700
X-Message-Number: 30
The way I read the path
(N1VG>APRS,K6SYV-10:}W4JRD-4>APW272,TCPIP*,N1VG*::K4REH-14), it originated
with W4JRD-4 running WinAPRS 2.72 and was explicitly routed through my
igate, N1VG (running aprsd), bound for K4REH-14. Now, I haven't used
WinAPRS, and I haven't really messed with APRS messaging much. Does the
user have to explicitly enter my IGATE in the path, or does it come from
somewhere else? And my transmission path only goes to the nearest WIDE,
K6SYV-10, and shouldn't really hit any other IGATEs. Except maybe for a new
one I saw pop up the other day, but that one hasn't been around long enough
to be the source of the problem.
----------------------------------------------------------------------
Subject: Re: APRS Network Protection.
From: "AE5PL Lists" <HamLists@ametx.com>
Date: Fri, 25 Jul 2003 14:36:49 -0500
X-Message-Number: 31
If the message is not acked and the WinAPRS continues to send the
message, the IGate should continue to gate the message. When you see
two of the same packets seconds apart, that indicates that the IGate
thinks this is a new message and sends it twice (not sure I like this,
but that is how it works). If N4REH-14 is a tracker, then no ack will
ever occur and WinAPRS should give up. This does not appear to be a
loop, but WinAPRS continuing to try to send the message.
IGates should never (unless they are also a digi) be specified in a
path. All transmissions and gating is done programmatically (IGates
maintain RF station lists, etc.). These packets were not explicitly
routed through your IGate. Your IGate sees K4REH-14 on RF; it sees the
messages for K4REH-14 from W4JRD-4 on APRS-IS; it gates to RF the
message in third-party format. Third-party packets with TCPIP* in the
header are not placed back into APRS-IS by IGates.
73,
Pete Loveall AE5PL
pete@ae5pl.net
----------------------------------------------------------------------
Subject: RE: Laptop alternative
From: Dennis Hudson <n2lbt@n2lbt.com>
Date: Fri, 25 Jul 2003 15:40:03 -0400
X-Message-Number: 32
Yes, it's unfortunate we didn't find this place when they were blowing
out the 266 processor model. A fine specimen for xastir.
----------------------------------------------------------------------
Subject: Re: Speaking of Bob's discussion of RF usage....
From: Bob Bruninga <bruninga@usna.edu>
Date: Fri, 25 Jul 2003 16:12:35 -0400 (EDT)
X-Message-Number: 33
On Fri, 25 Jul 2003, Dick Stanich wrote:
>>"Categorizing network behavior by the statistic of the number of
>>successful packets per hour can be completely missleading. A GREEN
>>indication of only 700 packets per hour throughput CAN mean a
>>lightly loaded network OR it can also indicate a totally swamped, totally
>>unreliable network suffering so many collisions that only 700 get
>>through."
>
>Ithink someone else will explain it better Bob. It measures all 'traffic'
>heard, doesn't determine whether packets are valid or not. Everytime the
>TNC detects a packet signal, corrupt or collision, it is counted. If the
>TNC lights up, it is traffic.
If it is watching the channel-busy LED, then it cannot be giving an output
in "packets per hour". Then the units would be "minutes per hour" of
channel busy. My suspicion is that it is only counting packets. and then
it can only count "valid" packets. (meaning received complete). A TNC
that hears wall-to-wall collisions gives no output except for a packet
that is 16 dB stgronger then any other and captures the FM channel.
Thus, it is an almost meaningless count of success when there is no info
regarding how many failures there are... Of course if you tie this to a
human ear that is listening to the channel at the repeater site, then
it is clearly easy to distinguish if the number means a lightly used
channel or a totally saturated one. But without the "ear" a value of 700
per hour means practically nothing...as far as network reliability to each
individual is concerned..
I'd love to learn that this assumption is wrong, but the only way to get
that info from a TNC is to have a hardwired connection to the channel-busy
LED at the digipeater. Even the CD line on the RS-232 port only reports
successful packets, it says nothing about channel collisions...
But I'd love to learn more...
thanks
bob
----------------------------------------------------------------------
Subject: Correct path to use on balloon
From: "Rochte, Robert" <rrochte@gpacademy.org>
Date: Fri, 25 Jul 2003 16:24:08 -0400
X-Message-Number: 34
The TT2 on our last tetroon used a path of RELAY,WIDE,WIDE - and I was
seeing packets on RF here in the Detroit area while it was over West
Virginia! I'm assuming this means that this path was overkill and did nasty
things to the network that day (sorry!)...
Any suggestions on how this should be set in the future?
Thanks.
73,
Robert
--
Robert Rochte
Director of Information Technology
The Grosse Pointe Academy
171 Lake Shore Road
Grosse Pointe Farms, MI 48236
Tel. +1 313-886-1221 x155
FAX +1 313-886-1418
www.gpacademy.org
----------------------------------------------------------------------
Subject: RE: Laptop alternative
From: "Mike Yetsko" <myetsko@insydesw.com>
Date: Fri, 25 Jul 2003 16:57:25 -0400
X-Message-Number: 35
About 2 years ago I sold a bunch of Lanner EM-520 boards to people on the
list here through the place I used to work.. They were actually not stock.
The stock EM-520 has VGA and TV out, these boards did NOT have the TV out
chip installed.
They weren't as 'self contained', but they still adaptable.
Pentium equivlent 180MHz processors, 4 com ports, 1 lpt, 1 floppy, 1 ATA
(as opposed to 2), DiskOnChip socket, PS/2 keyboard and mouse ports, Sound
port, 2 USB, 1 Ethernet, 1 ISA slot, 1 PC104 slot, VGA.
The ISA slot was actually one of these 'dual depth' slots that could also
expand to PCI and an ISA, although the adapter was hard to come by.
I was turning them out for well less than $100, including 2 SRAM sticks
(did I send 8 Meg or 16Meg, I can't remember).
You know, the place I worked still has a lot of them left.
Mike
----------------------------------------------------------------------
Read previous mail | Read next mail
| |
