[aprssig] IGATE message routing bug?

[Pete Loveall AE5PL Lists]

As stated in a recent thread, it is up to the IGate to determine what "local" is for messaging.  Unfortunately, the term "local" has been interpreted in multiple ways (hence the ambiguity in the aprs-is.net statement) over the years by various IGate authors.  Bob's digipeater paradigm of WIDEn-n has also changed over the years (note the major change when RELAY was supposed to be eliminated a number of years ago), the use of various numbers for the second n (WIDE2-1 for instance to only be digipeated by a single wide-area digipeater), and some digipeaters add/replace their callsign before the WIDEn-n and some don't, etc. makes hop counting a much bigger challenge than a simple "count the number of digipeaters and decremented hops".  So to "fix" this requires an agreement by -all- IGate authors on how to count hops (at least one is dead) and -all- digipeater sysops to adhere to Bob's paradigm (may require equipment replacement).

I am just laying out why this is not a simple "fix" and why the "fix" lies in the local IGates and local digipeaters, not the global servers, because each individual area in the world has a different definition of "local" for their individual RF requirements.  javAPRSSrvr's IGate adjunct allows the sysop to define how various local digipeaters behave so the sysop can better fine-tune "local" but I would wager to say most sysops simply install with defaults since things like ducting and changes in the local network can render such definitions invalid.

Hope this helps.

73,

Pete Loveall AE5PL
pete at ae5pl dot net



> -----Original Message-----
> From: Jim Alles
> Sent: Wednesday, November 16, 2016 2:00 AM
> Subject: Re: [aprssig] IGATE message routing bug?
> 
> "​
> Is it the failure of some local IGates or is it some kind of filtering issue in the
> APRS-IS.
> 
> Where do we fix this?"