[aprssig] Object wierdness

[Lynn W. Deffenbaugh (Mr)]

Earl Needham wrote:
> That sounds a LOT like what we occasionally see where someone will get home, shut off their mobile, and anywhere from several minutes to several hours later their "position" will move around the area and their tracker is SHUT OFF. 
>
> I don't think anybody has ever actually discovered why this happens.
>
>   
I discovered one way this can happen the other day when I inadvertently 
CAUSED it to occur.  I run the KJ4ERJ-2 IGate here in Palm Bay, FL.  At 
one point, I suspended a piece of the data path between the TNC and 
javAPRSsrvr for about 10-15 minutes.  When I discovered that and 
released the hold, 15 minutes worth of APRS packets were delivered into 
APRS-IS.  Since this was well past the default 30 second duplicate 
detection interval, they were all considered "new" and spread throughout 
the Internet.  If any of these stations had gone off-line during that 
interval, the delayed position update became the final position for that 
station, at least, until the station came live to change it again.

Ping-Ponging paths are also caused by these sorts of delays, but shorter 
than my 10-15 minute one (which I now know NOT to do!).  If a gated 
packet is bouncing through 3-4 digis on its way to a second IGate (or 
even bouncing back to the original gate), and arrives more than 30 
seconds after the original reception of the packet, the update will be 
applied.  If, during that bouncing time, the position was updated to be 
elsewhere, the station appears to go back in time.  I've observed these 
digi-induced delays to be upwards of 2-3 minutes in some busy APRS 
environments (where digis will buffer multiple received packets before 
bursting them back on the air when the channel is clear).

Anyway, this particular object (KD5LVH) does not appear to be this sort 
of event, but stemmed from a single beacon using that call entering 
APRS-IS.  I can't explain the lack of kill, but I believe I explained 
the original visibility of that callsign.

Lynn (D) - KJ4ERJ

> 7 3
> Earl
> KD5XB
>
> KG5LT wrote:
> I am helping N5AG figure out why an object still appears on aprs.fi.   
> The object name is KD5LVH.  It currently is "owned" by N5AG-5, a rf- 
> receive-only UI-View igate station (no RF xmit).  The current  
> timestamp tells me that it gets updated about every 22 minutes from  
> N5AG-5.  Earlier this afternoon in assisting N5AG in getting rid of  
> the object, I created an object on RF using xastir with KD5LVH name.   
> Sure enough on aprs.fi I was now the new "owner".  I waited a few  
> minutes, then using xastir, I deleted the KD5LVH object.
>
> Just now N5AG and myself were discussing this on the local repeater.   
> We both saw the KD5LVH object on aprs.fi and back at it's original  
> position.  N5AG-5 was now the owner of the object.  Just now, I see  
> where the object has moved to where it was the last time I had  
> created it and I am now the new owner.  Heh?  How can that be, I  
> haven't done anything with it on xastir for several hours.  This  
> starting to sound like the Twilight Zone...
>
> Question:  Does anyone have a clue as to why UI-View with igate  
> capability would be automatically creating this object?  N5AG as not  
> very familiar with UI-View, and neither am I.  Or, is this an issue  
> with aprs.fi that we both don't understand?
>
> It could be that *I* am clueless on how objects work.  I thought I  
> knew.  Now, I'm not so sure...
>
> 73 - Dale.  KG5LT
>
> KD5XB -- Earl Needham
> http://groups.yahoo.com/group/cw_bugs
> Quoting from the Coast Guard: ZUT
> Posted via Blackberry
> _______________________________________________
> aprssig mailing list
> aprssig at tapr.org
> https://www.tapr.org/cgi-bin/mailman/listinfo/aprssig
>
>