which tier servers farm does APRS.FI pull iGate data from

[N0AGI]

hello folks

I have been watching something weird w/ how my igate/digi (N0AGI) gated packets surface on APRS.fi. so let me explain what I'm noticing and perhaps someone here might be able to shed some light into this

so, my iGate/digi connects to the APRS IS servers farm - sometimes directly to first.aprs.net and other times to rotate.aprs.net on port 14580 and w/ my callsign/pwd creds. With this set up, my igate/digi has been gating RF packets to IS w/o any issues for at least 7+ months now.

I monitor my gated packets via an app I wrote in C#, that basically opens a read-only pull from first.aprs.net server and displays igated traffic data.  And, all of my igated packets show up on the pull report in realtime.  So, all that is goodness.

However, in the last 2 weeks, I have noticed that despite my igated packets end up in tier 1 servers, and they show that I was in the qAR "port of entry" into IS servers, APRS.FI site shows a different call sign for "port of entry".  For example, when I click on the "raw packets" link in the info page for a particular igated call sign, I see all of the packets that were igated by various igates.  Mine also show up there, but, not all of them.  In fact the most latest igated packets do NOT show up in the list. Thus rendering that I was never the igate to have gated that packet. When in reality my igated packet was the most recent

So, couple of questions:

1. has anyone else noticed this behavior. or is it just me and my igate going crazy :)

2. What is the source IS server that APRS.FI servers pull the IS gated data from. 

I think there might some latency issues on my end too. I don't know. Meaning, that while I igate a packet, my igated packet may not have reached the IS server quick enough for it to show and perhaps the other igate beat me to it.

anyways, wanted to check and see if anyone had any thoughts on this issue and any ideas I might and should explore to further discover the issue

thanks in advance all

73, N0AGI

[Heikki Hannikainen]

On Mon, 27 Apr 2015, N0AGI wrote:

> However, in the last 2 weeks, I have noticed that despite my igated packets end up in tier 1 servers, and they show
> that I was in the qAR "port of entry" into IS servers, APRS.FI site shows a different call sign for "port of
> entry".  For example, when I click on the "raw packets" link in the info page for a particular igated call sign, I
> see all of the packets that were igated by various igates.  Mine also show up there, but, not all of them.  In fact
> the most latest igated packets do NOT show up in the list. Thus rendering that I was never the igate to have gated
> that packet. When in reality my igated packet was the most recent
>
> So, couple of questions:
> 1. has anyone else noticed this behavior. or is it just me and my igate going crazy :)
> 2. What is the source IS server that APRS.FI servers pull the IS gated data from. 
>
> I think there might some latency issues on my end too. I don't know. Meaning, that while I igate a packet, my igated
> packet may not have reached the IS server quick enough for it to show and perhaps the other igate beat me to it.

aprs.fi gets the packets from all of the Tier2 hub servers. There is a
whole blog post about it:

http://blog.aprs.fi/2012/05/aprsfi-connected-directly-to-aprs2net.html

The APRS-IS servers filter duplicate copies of the same packet within 30
seconds. Each server only passes on the first copy of a packet it saw.
Depending on which server you connect to, you'll see different entry
igates, but you should get one copy of each packet.

If your stations shows up *sometimes* as the port of entry in raw packets,
then your igate is functional and there are also some other working igates
in your area. It's a matter of latency, and the network latency of the
server you're connecting to, and the latency of the tier2 hub server that
happens to sit between it and aprs2.net, if you're the first one or not.

If you're not always the first one, it's just a *good* sign - the network
works better if there is redudnancy brought in by multiple igates. Maybe
new ones have been set up in your area!

- Hessu

[Nagi Punyamurthula]

Thanks for sharing this link. Never knew about this posting.
I remember testing against the noam.aprs2.net server pool a few times and at that time, it seemed like the tier2 servers' connectivity was much slower than by connecting to on tier 1 pool. However, with that said, I have now changed my target to noam again. I will keep an eye and continue monitoring. Thus far it seems to be performing as I would expect to.

73, N0AGI-Nagi

[Nosey Nick]

On Tuesday, April 28, 2015 at 12:21:56 AM UTC-4, Heikki Hannikainen wrote:

aprs.fi gets the packets from all of the Tier2 hub servers. There is a
whole blog post about it:
http://blog.aprs.fi/2012/05/aprsfi-connected-directly-to-aprs2net.html

 

Hi Heikki / Hessu

Do you get CWOP from T2HUB as well, or do you have separate ingests for CWOP feeds as well?

Cheers
Nick

[Heikki Hannikainen]

On Tue, 28 Apr 2015, Nosey Nick wrote:

> On Tuesday, April 28, 2015 at 12:21:56 AM UTC-4, Heikki Hannikainen wrote:
> aprs.fi gets the packets from all of the Tier2 hub servers.
> There is a
> whole blog post about it:
> http://blog.aprs.fi/2012/05/aprsfi-connected-directly-to-aprs2net.html
>

> Do you get CWOP from T2HUB as well, or do you have separate ingests for CWOP
> feeds as well?

Separate connections to each CWOP server.

- Hessu