Distance bug?

[Remi Hutin F6CNB]

Hi Bela,

Thanks for this very nice program.

I am using it on a Raspberry Pi with a soundmodem as a received only igate in Texas>

I saw a completely wrong distance reporting on the status page

KE5WVC-10 is reported to be at 3786km from me when he is only at 167km

73 de Remi W5/F6CNB

See below the status page and the raw packets:

DIXPRS STATUS PAGE

SETUP

Callsign:
F6CNB-14

Version:
 DIXPRS 2.2.2 05-09-2012
 
Platform:
Linux 3.10.25+ (armv6l), Python 2.7.3

Position:
29.8547N 96.5274W

Owner:
Remi,W5/F6CNB

Port(s):
144.390MHz

SYSTEM

Time:
 07-05-2014 08:32:12z

Started:
 06-05-2014 23:40:32z

Uptime:
 0 days 8 hours 51 mins

INFO
 
Positions stored:
  212
 
Heard direct in last hour:
 AB5CC-2 AE4S-1 AI5TX-7 K5FRC-7 K5GVR-10 K5KRK-1 K5PAL-1 KA5WMY-1 KC5EZZ-4 KD5QZD-6 KE5WVC-1 VA3FES-14 W4ZFH-2 W5BEC-5 W5BEQ-1 W5GC-2 WA5LNL-3 WD5IYT-2 WR5AAA-1

IGATE counters:
MSG_CNT=0 LOC_CNT=73 DIR_CNT=19 RF_CNT=127
 
DX (1h)
 K5GVR-10 07-05-2014 08:02z 724.5 km
AB5CC-2 07-05-2014 08:30z 458.0 km

DX (24h)
 KE5WVC-1 07-05-2014 08:29z 3785.8 km
K5GVR-10 07-05-2014 08:02z 724.5 km

DX (all)
 KE5WVC-1 07-05-2014 08:29z 3785.8 km
K5GVR-10 07-05-2014 08:02z 724.5 km

 Page generated in 5.642 seconds

Raw packets from aprs.fi (CDT is 5hr off from GMT/UTC)

2014-05-07 03:27:04 CDT: KE5WVC-1>APN383,WIDE2-2,qAR,F6CNB-14:!!000000FF032809722794032E01E3----0006029900020000
2014-05-07 03:27:06 CDT: KE5WVC-1>APN383,WD5IYT-2*,WIDE2-1,qAR,WB5AOH:!!000000FF032809722794032E01E3----0006029900020000 
2014-05-07 03:29:34 CDT: KE5WVC-1>APN383,WIDE2-2,qAR,F6CNB-14:!3102.63N/09735.08W_PHG2260/W_ KE5WVC-1,BELTON,TX, A CTARN Station...
2014-05-07 03:32:05 CDT: KE5WVC-1>APN383,WIDE2-2,qAR,F6CNB-14:!!000000FF032809722794032E01E3----0006029900020000

[Bela Markus]

Hi Remi,

I can't tell you why KE5WVC-1 has such distance on the DX list. What is the distance displayed on the terminal when it is heard? Are others on the DX list and the monitor have the correct distance?

Regards... Bela

--
Azért kapta ezt az üzenetet, mert feliratkozott a Google Csoportok „dixprs” csoportjára.
Az erről a csoportról és az ahhoz kapcsolódó e-mailekről való leiratkozáshoz küldjön egy e-amailt a(z) dixprs+un...@googlegroups.com címre.
További lehetőségekért látogasson el ide: https://groups.google.com/d/optout.

A(z) üzenetben nem található vírus.
Ellenőrizte: AVG - www.avg.com
Verzió: 2014.0.4570 / Vírus adatbázis: 3931/7451 - Kiadás dátuma: 2014.05.06.

[Manos Darkadakis]

Hi all

I have a similar issue some time ago but haven’t paid much attention to it. I have now removed the database files and I will keep an eye and report if a similar situation occurs…

Dixprs works flawlessly for many months now with the only addition of a small watchdog script to check for hardware lockups and reboot if that happens. So practically is unattended system. Now looking to make it “green” under solar power….

 

Regards Manos SV1IW

 

---

[Bela Markus]

Remi

my advice is the same, stop system, delete database and restart. It reminds me to create a tool to delete selected item(s) from the DX list. On my systems or systems around me fals distance was always due to false position sent by the other station due to GPS, wrong configuration ot buggy software or tracker.

Regards... Bela

Verzió: 2014.0.4570 / Vírus adatbázis: 3931/7462 - Kiadás dátuma: 2014.05.08.

[Ray Wells]

Hi Bela,

I have a question for your features list, prompted by a question on the
RPi list about a command line aprs program for mobile work.

Can dixprs be made to accept input from a gps via a port looking at gpsd
and beacon that information as a not-so-dumb tracker?

Regards,
Ray vk2tv

[Bela Markus]

Hi Ray,

it is a work in progress. There are two phases planned. First to use GPS
to determine position. Application is a mobile digi/gateway for field
events. Next to implement a real tracker. No ETA yet, but hopefully it
will come.

Regards... Béla

[Remi Hutin F6CNB]

Bela,

I stopped it , deleted the database and restarted it.
I'll keep an eye on it.

Several stations had wrong distance. They were all far (several degrees) on the northwest on me.
(I am in Texas with negative longitude). I'll try to find the root cause , if it is happening again.
I need very good propagation to catch these stations.

73 de Remi W5/F6CNB

[Bela Markus]

Remi,

if you see wrong distance, please make a screen capture of the console and mail to compare it with raw data on aprsi.fi We have many dixprs running in HA and wrong distance is always casued by false transmitted position when I analysed. But lets see when we get data.

What is happaning more often, reporting a station heard directly which was not heard for sure. It is caused by wrong AX.25 headers created by buggy digis.

Regards... Bela

Verzió: 2014.0.4570 / Vírus adatbázis: 3931/7470 - Kiadás dátuma: 2014.05.10.

[Remi Hutin F6CNB]

Bela,

I think I found the root cause of the wrong distance.

All the stations with wrong distances are transmitting messages like the following one:

APN383,WIDE2-2,qAR,F6CNB-14:!!000000FF032809722794032E01E3----0006029900020000

Notice the double !! at the beginning of the message. It is interpreted by dixprs as a regular position message !DDmm ......

which results in a wrong position and then a wrong distance.

I did not find the meaning of these kind of messages in the APRS specifications but several stations are using it around here.

Could you add a check in the software that the format after the first ! is correct>

Thanks  

73 de Remi W5/F6CNB

[Remi Hutin F6CNB]

This is the weather message from an Ultimeter 2000 weather station.

Remi

[Bela Markus]

Hi Remi

thanks for the info, will check the code.

Regards... Bela, HA5DI

[Lynn W Deffenbaugh (Mr)]

Chapter 12 of aprs101.pdf.

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

[Bela Markus]

Hi

issue fixed in dixlibaprs.py, see git repo at https://github.com/ha5di/DIXPRS

Just replace dixlibaprs.in your installation. Feedback welcome.

Regards... Béla, HA5DI

[Remi Hutin F6CNB]

Bela,

Thanks for the patch. I ran it on my US APRS node and it worked perfectly.

I am now receiving the ultimeter 2000 stations without a distance error.

The patch has fixed the distance bug.

73 de Remi W5/F6CNB

[Remi Hutin F6CNB]

After 6 days, No distance errors

Thanks again for the patch.

Remi