[Aprsd-users] Re: "local" Internet to RF gating

[Dennis H. N. <n2...@n2...>]

On 7/25/01 6:32 PM, "Hamish Moffatt" <ha...@cl...> wrote:

> On Wed, Jul 25, 2001 at 05:24:06PM -0400, Dennis Hudson, N2LBT wrote:
>> It is in reference to the internet to RF messaging. The problem I am having
>> is that because of either propagation or improper path selection, my igate
>> hears a packet from distant stations and attempts to pass traffic to that
>> station as if it was local. The message will in most cases never reach its
>> party as those stations are far beyond my two digi hop. In most cases those
>> stations are served locally by another Igate or are actually in direct
>> contact with the messaging station on RF.
> 
> Yes, that's how it is supposed to work. (If it has heard a station locally
> within the history time period, it will send messages for that station
> from Internet to RF.) How do you want this changed?
> 
> 
> 73
> hamish

Thanks for replying Hamish, I realize that the code is working, but
circumstances have presented themselves frequently that cause the original
plan to appear to malfunction in our area.

As far as I know I perceive the following to be true, and therefore am
wondering if it is worth looking at. I will elaborate on our local APRS LAN
and what type of packets cause trouble here. In New England we have the
problem of many bumps or hills (we call them mountains) at similar elevation
which force us to use many digis where others can get away with one "super
digi". There are many APRS populated areas separated by only 60-70 miles and
a digi or two for each area. The IGates have paths that cover the local area
usually a path of DIGI1,WIDE. Although I envisioned an IGate for each of
these populated areas, they have not caught on as quickly as I hoped.

The packets/scenario I see as a problem is a mobile or fixed station that
exceeds my UI TNC/IGate path. What I call a DX station 3,4,5,6,7 hops away
is heard by my IGate and aprsd automatically tags that as a local station.
If aprsd was "smart" and adjusted the path to actually send traffic for that
station with DIGI1,DIGI2,DIGI3,DIGI4 I wouldn't have as much as a problem
with this. Better yet, not RF gate messages for stations that have 5% or
less chance of getting the traffic.  But in most cases that gated packet
will probably will get passed by another IGate, or even directly reach its
party on RF. I realize the offending station is really to blame, and they
should only be using paths that are necessary. Unfortunately it's very
difficult to remind each an every operator about their paths on an everyday
basis. And even then "DX" packets reach the local LAN due to propagation or
experimentation by a APRS station. I would prefer the IGate to try to
recognize the impossibility of trying to pass traffic to a station 5 hops or
hundreds of miles away.

I assume that a station is considered local if it is heard on any of aprsd's
ports. What is the TTL for a station heard on one of these ports before it
not considered a "local" anymore? Should I run a script to restart the IGate
every few hours to overcome these problems? I often leave the aprsd code
running for hundreds of hours at a time without rebooting. Mostly because I
don't know how to program scripts to start and stop programs and the
computer.


-- 
Dennis, N2LBT
Sysop aprsdALY IGate
http://www.n2lbt.com
n2...@n2...