[aprssig] APRS-IS and metadata

[Lynn W Deffenbaugh (Mr)]

I suspect the "metadata" is something like what I recently started 
seeing attached to YO2MAS-7's apparently LoRa packets by the YO2MD-11 
and YO2MAS-11 IGates. This is a serious breach of APRS-IS standards, 
specifically "No modification of the TNC2 format line should be made 
except to add ,qAR,IGATECALL to the end of the path" 
(https://www.aprs-is.net/IGating.aspx). Any modification of or addition 
to the received packet content breaks the duplicate filter of the 
APRS-IS, not to mention this particular modification will cause the 
appended text to appear as the comment string on the originating station.

> 2023-10-10 06:19:40 EDT: 
> YO2MAS-7>APLT00,WIDE1-1,Q03K8,qAR,YO2MAS-10:!4611.63NL02118.17Ea234/000/A=000837LoRa 
> Tracker -  Bat.: 4.13V - I:  0mA !wp/! 2023-10-10 06:19:41 EDT: 
> YO2MAS-7>APLT00,WIDE1-1,qAO,YO2MAS-11:!4611.63NL02118.17Ea234/000/A=000837LoRa 
> Tracker -  Bat.: 4.13V - I:  0mA !wp/:> YO2MAS-11:>Last RX: SNR=8.75dB 
> RSSI=-72dB [Rate limited (< 5 sec)] 2023-10-10 06:19:41 EDT: 
> YO2MAS-7>APLT00,WIDE1-1,qAO,YO2MD-11:!4611.63NL02118.17Ea234/000/A=000837LoRa 
> Tracker -  Bat.: 4.13V - I:  0mA !wp/:> YO2MD-11:>Last RX: SNR=8.25dB 
> RSSI=-73dB [Rate limited (< 5 sec)]

And just in case the IGate software authors get the inspiration for more 
changes, this is also in the APRS-IS IGate specification: "IGates must 
not modify paths of packets gated to APRS-IS except to append 
,qAR,IGATECALL" (https://www.aprs-is.net/IGateDetails.aspx).   If you 
notice, some of the "good" (black) raw packets at aprs.fi have an 
addition Qnnnnn element before the qAR construct, coming through the 
YO2LIS-10, -11, and -12 IGates according to the qAR which may or may not 
be trustworthy.

Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32

PS. https://aprs.fi/?c=raw&call=YO2MAS-7&limit=1000&view=normal



On 10/9/2023 9:08 PM, Nick VA3NNW wrote:
>> When I talk about metadata I mean metadata stored into the backend databases, that should be supplided by the iGates. Maybe it would be time to define a new mechanism now that at least LoRa-APRS is reasonably young and there are still few developers.
>
> What are the "backend databases" to which you refer?
>
> APRS-IS is essentially just a giant stream of de-duped TNC2-format 
> lines. No storage, just one continuous "live" stream.
>
> There's a hierarchy of servers, which all send all the packets to each 
> other, up and down the tree, and around in loops, except thanks to the 
> de-duping, loops never actually cause loops in the actual data itself.
>
> Now possibly SOME people take that stream and put it into databases 
> (aprs.fi, aprsdirect.de spring to mind, probably dozens of others), 
> but you're not feeding all of them, you're feeding APRS-IS which in 
> turn is feeding all of those (and many thousands of other clients, all 
> somewhat alike, and all expecting TNC2 data). APRS-IS itself has no 
> "backend databases" as such.
>
> You would presumably need to find some way to add your metadata to 
> TNC2-format lines... whilst also not breaking that de-duplication BTW, 
> and also not breaking any other servers or clients that are expecting 
> a stream of TNC2-format lines.
>
> Nick VA3NNW
> -- 
> "Nosey" Nick Waterman, VA3NNW/G7RZQ, K2 #5209, IRLP #2410 on VE3RBM
> use Std::Disclaimer;sig at noseynick.net
> Do Passwords PROPERLY! ... seehttps://noseynick.net/vault.html
> Those who live in glass houses...shouldn't.
>
> _______________________________________________
> aprssig mailing list
> aprssig at lists.tapr.org
> http://lists.tapr.org/mailman/listinfo/aprssig_lists.tapr.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.tapr.org/pipermail/aprssig_lists.tapr.org/attachments/20231010/f6278816/attachment.html>