[aprssig] Unique Callsign-SSIDs PLEASE

[AE5PL Lists]

Time for my annual rant about unique callsign-SSIDs and multiple
simultaneous connections to APRS-IS.  Whether we like it or not, APRS is
a world-wide network transported on an archaic but effect transport
mechanism called APRS-IS.  This world-wide interconnect brings some
baggage along with it, most important of which is the requirement for
-every- station to use an unique callsign-SSID combination.  Use of
duplicate callsign-SSIDs causes dropped packets, improper gating of
packets to RF, multiple responses to messages, etc.  The biggest issue I
tend to see is when someone sets up an RF station and a weather station
using the same exact callsign-SSID (no SSID is actually a SSID of zero).
However, there are other instances such as software configured to make
multiple simultaneous connects to APRS-IS thereby having the appearance
of multiple stations with the same callsign-SSID.

PLEASE review your station and software configurations to ensure that
there are -no- duplicate callsign-SSIDs being used.  Remember that if
you are not transmitting directly to RF (a weather station connected
only to the Internet, for instance), that station can use a non-numeric
SSID.  Even when gated to RF, those stations will be interpreted
correctly even by the D-700s and D-7s.  Do not use zero (0) as a SSID.
This can confuse some software, especially if connected directly to a
TNC.  Use no SSID instead (just a callsign).
 
For those of you running servers or software with multiple simultaneous
connection capabilities, please restrict your configuration to a single
upstream (outbound) connection.  Due to the primitive techniques that
must be employed to reduce loops, having multiple connections actually
decreases reliability by introducing dropped packets due to those
packets being considered looped packets.

Summary: every piece of software and every station MUST have unique
callsign-SSID combinations; every piece of Internet-connected software
MUST use only one connection at a time to the rest of APRS-IS.

73,

Pete Loveall AE5PL
www.ae5pl.net