[aprssig] Delayed Packets

[Wes Johnston]

Nothing difinitive, but we had the same problem of delayed IGATEd
packets, and it was related to one guy in our area running UI-View.

I was able to pull data from findu and see duplicate packets that were
delivered out of sequence by 5 to 10 minutes.

When he shut down his UIview station, the problem went away.
Wes

Stan Coleman N0YXV wrote:
> We have what I think is a unique thing happening in our town. On several
> occasions now we've had packets show up minutes if not hours later after
> they were transmitted. On Saturday we had a mobile that would show up on one
> side of town then back at his home. I guess (from what I was told) that it
> ping ponged across town several times. I did look up my RF log from Saturday
> morning and didn't find ANY duplicates. I didn't see this event but I did
> experience the event in the next paragraph.
> 
> I had an occasion where I went to work and well over an hour after I got to
> work I saw a packet coming in from my car. This was a little bit worrisome
> but it was showing still in the packing lot. After the third packet from the
> parking lot I got up and went out to check my car. Nobody had been in my car
> and it wasn't on. Those packets turned out to be repeats that had occurred
> well over an hour before.
> 
> In town we only have one Digi (that I know of) and that is WY0F. We do
> however have three full time IGates and 2 others that can be IGates at any
> one time. If you pull up our town that's a huge number of IGates in a very
> small area. Now here is my question, "Is it possible for all of these IGates
> to be transmitting packets to that APRS network (Internet) at slightly
> different times?" (See my guess below for a better explanation.)
> 
> Let me know if I'm really off base but here is my guess as to what happens.
> Packet A gets IGated at two or more locations. Location one is using a Tier
> 2 server and location one is using a Tier 1 server (different servers in any
> case). If the packets taking the longest path gets buffered anywhere along
> the path that might be enough delay to cause the repeats. The packet going
> through Tier 2 gets delayed enough to look enough different than the packet
> going through Tier one that the system thinks they are two different packets
> and puts both into the system. Thus making the vehicle appear to bounce
> between two locations. Since we basically are using what amounts to telnet
> packets and not a streaming system some routers put telnet packets at the
> lowest end of their queue systems.
> 
> Almost everyone in town is IGating using UI-View and I'm (N0YXV-1) IGating
> with Javaprssrvr. IGates are WY0F, KC0MWM, N0YXV-1, Digi is WY0F. The whole
> system is WIDEn-N compatible.
> 
> If my theory is true I think the only fix would be to make sure that all
> IGates in a local LAN area use the same internet server.
> 
> _______________________________________________
> aprssig mailing list
> aprssig at lists.tapr.org
> https://lists.tapr.org/cgi-bin/mailman/listinfo/aprssig
> 
> 
> 
>