[aprssig] I-gate loops for telemetry support text messages
Pete Loveall AE5PL Lists
hamlists at ametx.com
Wed Sep 22 19:09:57 EDT 2021
Yes. Both 1 & 2. IGates should not gate to RF message packets that are "self-addressed" (should only be Telemetry Messages but might be others). IGate gate range should be based on hops and should not gate beyond "local" (however you define that and I know that it is difficult with the various implementations of the WIDEn-n paradigm).
Hope this helps.
73,
Pete Loveall AE5PL
pete at ae5pl dot net
-----Original Message-----
From: aprssig <aprssig-bounces at lists.tapr.org> On Behalf Of Andrew Pavlin via aprssig
Sent: Wednesday, September 22, 2021 4:45 PM
To: TAPR APRS Mailing List <aprssig at lists.tapr.org>
Subject: [aprssig] I-gate loops for telemetry support text messages
Greetings.
I just discovered an interesting "loop" caused by I-gating. My Tx I-gate received a text message from APRS-IS that was originally sent by a digipeater that I received both by Internet and by long-range digpeater hop on RF. So, I've heard the station on RF (but not _local_ RF, because there were one to three digipeaters in the path by the time I received it on RF). It is one of the self-addressed text messages that go with telemetry messages to say how to interpret the telemetry (EQNS, BITS, UNIT, PARM).
So, my Tx I-gate, having received from the Internet a text message addressed to a station it has heard over RF, proceeded to transmit the text message to RF (but without enough digipeat aliases to actually reach the station).
This sounds like a bug in my I-gate. So is the correct solution:
1. never Tx I-gate self-addressed text messages.
2. never Tx I-gate to RF stations that require more hops than I allow on my Tx I-gate digipeat path (i.e., for Tx I-gates that can put a WIDE1-1 or WIDE2-1 on their retransmissions). For the common case of no digipeat path for Tx I-gating, the RF station would have to be heard directly (not by digipeater) by the I-gate.
So how should I fix my I-gate?
Andrew, KA2DDO
author of YAAC ("Yet Another APRS Client")
More information about the aprssig
mailing list