Announcement

Collapse
No announcement yet.

Linux feeder software for Flightradar24 (Old software)

Collapse
This topic is closed.
X
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Originally posted by FRC View Post
    Hello,

    Does anyone is having problem with statistics page again (T-Feeder) with Rpi ???
    I am using fr24feed_arm-rpi_242 with latest Dump1090 (MalcolmRob fork) and latest Rpi firmware.
    I am using a ssh conection with Rpi and if i run the feeder in the foreground it seems to work ok. But if i send it to the background (&) , the statistics page does not seem to be updated, even when status shows "Online".

    Regards,

    Marcelo.
    still work with screen? I assume RPi includes screen?

    often in linux it is far better to run things that dont have a daemon setting to fork in background, to use screen rather than ampersand

    Edited: I have seen many programs stop when &'s, an others I've seen become unpredictable in behaviour... just a thought...
    Last edited by Ressy; 2014-06-21, 01:53. Reason: afterthought

    Comment


    • Originally posted by FRC View Post
      Getting nuts here Itīs late, iīm tired and iīll give up for today.

      The only message that i noticed that donīt show on log in deatched mode is:

      [e][stats] Cached feed_id=XXXX for network T.

      This could be an account problem too, i really donīt know.

      Any ideas ????

      Maybe i should stop feeding for some days and try again ....
      This is what I get

      Command - this is how I normally run it
      ./fr24feed_arm-rpi_242 --fr24key=xxxxxxxxxxxxxxxxxxxx --bs-ip=127.0.0.1 > /tmp/fr24feed.log &

      Log file starts
      Code:
      [i]FR24Feed v242 - built on May  7 2014/13:05:06
      [i]Downloading configuration...OK
      [i]Parsing configuration...OK
      [c]Interval: 5s
      [c]Latitude: xxxxxxxxxxxxxx
      [c]Longitude: xxxxxxxxxxxx
      [c]GND: YES
      [c]NonADSB: YES
      [i]Source defined as Basestation
      [b]connecting to 127.0.0.1:30003
      [e]could not connect to BS
      [i]waiting 5 seconds
      [n]defined 1 server
      [n]EGBB19@83.140.21.66:8099/UDP
      [n]connecting
      [i][stats]Stats thread started
      [n]connected
      [n]switching to UDP
      [n]working
      [i]Source defined as Basestation
      [b]connecting to 127.0.0.1:30003
      [b]connected
      [b]working
      [i]Data feed time difference OK abs(02:07:07.801 - 02:07:07.772)=0
      [i]sent 14 planes in 1 packets
      [i]sent 14 planes in 1 packets
      [i]sent 14 planes in 1 packets
      command - as above no log file
      Code:
      root@raspberry-pi:~# ./fr24feed_arm-rpi_242 --fr24key=xxxxxxxxxxxxxxxxxxxx --bs-ip=127.0.0.1 &
      [1] 4474
      root@raspberry-pi:~# [i]FR24Feed v242 - built on May  7 2014/13:05:06
      [i]Downloading configuration...OK
      [i]Parsing configuration...OK
      [c]Interval: 5s
      [c]Latitude: xxxxxxxxxxxxx
      [c]Longitude: xxxxxxxxxxxx
      [c]GND: YES
      [c]NonADSB: YES
      [i]Source defined as Basestation
      [b]connecting to 127.0.0.1:30003
      [b]connected
      [b]working
      [n]defined 1 server
      [n]EGBB19@83.140.21.66:8099/UDP
      [n]connecting
      [i][stats]Stats thread started
      [i]Data feed time difference OK abs(06:11:16.062 - 06:11:16.067)=0
      [n]connected
      [n]switching to UDP
      [n]working
      [i]sent 79 planes in 2 packets
      [i]sent 82 planes in 3 packets

      Just running as a interactive putty task
      Code:
      root@raspberry-pi:~# ./fr24feed_arm-rpi_242 --fr24key=xxxxxxxxxxxxxxx --bs-ip=127.0.0.1
      [i]FR24Feed v242 - built on May  7 2014/13:05:06
      [i]Downloading configuration...OK
      [i]Parsing configuration...OK
      [c]Interval: 5s
      [c]Latitude: xxxxxxxxxxx
      [c]Longitude: xxxxxxxxxx
      [c]GND: YES
      [c]NonADSB: YES
      [i]Source defined as Basestation
      [b]connecting to 127.0.0.1:30003
      [n]defined 1 server
      [n]EGBB19@83.140.21.66:8099/UDP
      [n]connecting
      [b]connected
      [b]working
      [i][stats]Stats thread started
      [i]Data feed time difference OK abs(06:12:38.308 - 06:12:38.312)=0
      [n]connected
      [n]switching to UDP
      [n]working
      [i]sent 82 planes in 3 packets
      [i]sent 76 planes in 2 packets
      [i]sent 76 planes in 2 packets
      [i]sent 80 planes in 2 packets
      I got one error in the top block - could not connect to base station

      All blocks started the 'stats' thread.

      Maybe download and re-install the feeder (it was re-issued wasn't it???), if that doesn't do it ask support ... they may have a debug switch to turn on that saves more info.
      Last edited by peterhr; 2014-06-21, 06:24. Reason: munged the location in the logs

      Comment


      • Just to follow up the information, my log output from foreground run

        Code:
        [i]FR24Feed v242 - built on May  7 2014/13:05:06
        [i]Downloading configuration...OK
        [i]Parsing configuration...OK
        [c]Interval: 5s
        [c]Latitude: XXXXXXX
        [c]Longitude:XXXXXXX
        [c]GND: YES
        [c]NonADSB: YES
        [i]Source defined as Basestation
        [b]connecting to localhost:30003
        [i][stats]Stats thread started
        [n]defined 1 server
        [n]SBFC2@83.140.21.66:8099/UDP
        [n]connecting
        [b]connected
        [b]working
        [i]Data feed time difference OK abs(19:49:37.480 - 19:49:37.473)=0
        [n]connected
        [n]switching to UDP
        [n]working
        [i]sent 4 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 4 planes in 1 packets
        [i]sent 3 planes in 1 packets
        [i]sent 3 planes in 1 packets
        [i]sent 4 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [n]pinging the server
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 5 planes in 1 packets
        Now, the log from detatched run

        Code:
        [i]FR24Feed v242 - built on May  7 2014/13:05:06
        [i]Downloading configuration...OK
        [i]Parsing configuration...OK
        [c]Interval: 5s
        [c]Latitude: XXXXXXX
        [c]Longitude: XXXXXXX
        [c]GND: YES
        [c]NonADSB: YES
        [i]Source defined as Basestation
        [b]connecting to localhost:30003
        [n]defined 1 server
        [n]SBFC2@83.140.21.66:8099/UDP
        [n]connecting
        [i][stats]Stats thread started
        [b]connected
        [b]working
        [i]Data feed time difference OK abs(21:00:17.513 - 21:00:17.496)=0
        [n]connected
        [n]switching to UDP
        [n]working
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 7 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [n]pinging the server
        [i]sent 5 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 5 planes in 1 packets
        [i]sent 6 planes in 1 packets
        But after so many tests, i did noticed that both executions was not updating the statistics page.
        So the idea that was difference between fr an bg was not confirmed.

        I tried today with another environment. A P3 pc + Debian Squeeze + i386 feeder client and it seems to work ok. In fact
        i was using this equip. before trying the Rpi.

        The output log from PC.

        Code:
        [i]FR24Feed v242 - built on Apr 23 2014/10:19:07
        [i]Downloading configuration...OK
        [i]Parsing configuration...OK
        [c]Interval: 5s
        [c]Latitude: XXXXXXX
        [c]Longitude: XXXXXXXX
        [c]GND: YES
        [c]NonADSB: YES
        [i]Source defined as Basestation
        [b]connecting to localhost:30003
        [b]connected
        [b]working
        [n]defined 1 server
        [n]SBFC2@83.140.21.66:8099/UDP
        [n]connecting
        [i][stats]Stats thread started
        [i]Data feed time difference OK abs(10:01:18.023 - 10:01:17.977)=1
        [n]connected
        [n]switching to UDP
        [n]working
        [i]sent 8 planes in 1 packets
        [i]sent 9 planes in 1 packets
        [i]sent 8 planes in 1 packets
        [i]sent 7 planes in 1 packets
        [i]sent 8 planes in 1 packets
        [i]sent 7 planes in 1 packets
        [i]sent 7 planes in 1 packets
        [i]sent 7 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 6 planes in 1 packets
        [i]sent 7 planes in 1 packets
        [i]sent 9 planes in 1 packets
        [i]sent 8 planes in 1 packets
        [n]pinging the server
        [i]sent 6 planes in 1 packets
        [i]sent 7 planes in 1 packets
        [i]sent 7 planes in 1 packets
        [i]sent 8 planes in 1 packets
        I will try with Rpi another day. Just not sure what is going on.

        ----
        Linux raspberrypi 3.12.21+ #689 PREEMPT Wed Jun 11 21:45:12 BST 2014 armv6l
        Uptodate Raspian distro
        ----

        Comment


        • It's interesting that there is a separate stats thread and that seems to be the part not working on your Pi.

          It might be worth trying the experimental feeder http://forum.flightradar24.com/threa...ting-(Win-RPi) , It seems to work OK but I'm not using it until someone says how to get a local coverage map out of it (like you can get with dump1090 on port 8080 or adsbscope on a windows PC

          Comment


          • Itīs working.
            Before trying again today, i did an apt-get update / upgrade. Thatīs all.
            Same Dump1090 and feeder version.

            I just noticed that the statistics page takes 1 hour (avg.) to get updated.

            ---
            Linux raspberrypi 3.12.22+ #691 PREEMPT Wed Jun 18 18:29:58 BST 2014 armv6l
            Up to date Raspian distro
            ---

            Comment


            • Originally posted by FRC View Post
              I just noticed that the statistics page takes 1 hour (avg.) to get updated.
              That is correct, results are cached for up to an hour.

              Comment


              • Hi guys, I have following setup:

                Raspberry Pi in an outdoor box with RTL-SDR dongle. Dump1090 was running on this machine and fr24 feeder was running on my server in the house.
                I wanted to lower the load on raspi so I moved dump1090 on the same machine which fr24 feeder runs (data transfer from rpi using nc and capture using rtl_sdr). I can see output correctly in dump1090.

                However fr24 feeder has some problems with time difference. What exactly these time stamps in the output mean? the first one is source data timestamp and the second one is fr24 feeder current timestamp?
                Any ideas how to debug this?
                Anybody running similar setup to mine?

                Code:
                [i]FR24Feed v242 - built on May  7 2014/11:42:36
                [i]Downloading configuration...OK
                [i]Parsing configuration...OK
                [c]Interval: 5s
                [c]Latitude: 50.0600
                [c]Longitude: 14.2869
                [c]GND: YES
                [c]NonADSB: YES
                [i][stats]Stats thread started
                [i]Source defined as Basestation
                [b]connecting to localhost:30003
                [n]defined 1 server
                [n]LKPR14@83.140.21.66:8099/UDP
                [n]connecting
                [b]connected
                [b]working
                [e]Data feed time difference too big abs(11:48:40.256 - 11:58:37.177)=597
                [e]Data feed time difference too big abs(11:48:40.289 - 11:58:37.180)=597
                [e]Data feed time difference too big abs(11:48:40.307 - 11:58:37.181)=597
                [n]connected
                [n]switching to UDP
                [e]Data feed time difference too big abs(11:48:40.302 - 11:58:37.444)=597
                [n]working
                [e]Data feed time difference too big abs(11:48:40.307 - 11:58:37.576)=597
                [e]Data feed time difference too big abs(11:48:40.252 - 11:58:37.836)=597
                [e]Data feed time difference too big abs(11:48:40.263 - 11:58:37.837)=597
                [e]Data feed time difference too big abs(11:48:40.275 - 11:58:37.838)=597
                [e]Data feed time difference too big abs(11:48:40.275 - 11:58:37.839)=597
                [e]Data feed time difference too big abs(11:48:40.288 - 11:58:37.840)=597
                [e]Data feed time difference too big abs(11:48:40.258 - 11:58:37.903)=597
                [e]Data feed time difference too big abs(11:48:40.281 - 11:58:37.905)=597
                [e]Data feed time difference too big abs(11:48:40.248 - 11:58:37.968)=597
                [e]Data feed time difference too big abs(11:48:40.280 - 11:58:37.971)=597
                [e]Data feed time difference too big abs(11:48:40.275 - 11:58:38.036)=598

                Comment


                • The easy solution would be to make sure your time is accurate and the same on both the Pi and the server. Seems you have a 10 minute "error" there.

                  Comment


                  • yep, your server time is off by 10 minutes.

                    ( I'd run it all on the Pi, and not have a time sync concern )

                    Comment


                    • HermanZA: what is interesting is that the first time stamp does not vary a lot, however the second one runs forward - seems that the first time stamp is about the time I started RTL dump process on rpi.

                      peterhr: I have some other functions running on Rpi and I wanted to off-load all functionality which does not need to run on Rpi directly...

                      Actually both machines are in good time sync. Reported difference is not real. As seen from the output, it is constantly increasing:
                      Code:
                      [i]FR24Feed v242 - built on May  7 2014/11:42:36
                      [i]Downloading configuration...OK
                      [i]Parsing configuration...OK
                      [c]Interval: 5s
                      [c]Latitude: 50.0600
                      [c]Longitude: 14.2869
                      [c]GND: YES
                      [c]NonADSB: YES
                      [i][stats]Stats thread started
                      [i]Source defined as Basestation
                      [n]defined 1 server
                      [b]connecting to localhost:30003
                      [n]LKPR14@83.140.21.66:8099/UDP
                      [n]connecting
                      [b]connected
                      [b]working
                      [n]connected
                      [n]switching to UDP
                      [n]working
                      [e]Data feed time difference too big abs(14:04:01.578 - 14:06:47.694)=166
                      [e]Data feed time difference too big abs(14:04:01.561 - 14:06:47.758)=166
                      [e]Data feed time difference too big abs(14:04:01.534 - 14:06:47.821)=166
                      [e]Data feed time difference too big abs(14:04:01.528 - 14:06:47.887)=166
                      [e]Data feed time difference too big abs(14:04:01.535 - 14:06:47.954)=166
                      [e]Data feed time difference too big abs(14:04:01.573 - 14:06:48.157)=167
                      [e]Data feed time difference too big abs(14:04:01.578 - 14:06:48.158)=167
                      [e]Data feed time difference too big abs(14:04:01.560 - 14:06:48.221)=167
                      [e]Data feed time difference too big abs(14:04:01.577 - 14:06:48.223)=167
                      [e]Data feed time difference too big abs(14:04:01.561 - 14:06:48.287)=167
                      [e]Data feed time difference too big abs(14:04:01.526 - 14:06:48.350)=167
                      [e]Data feed time difference too big abs(14:04:01.533 - 14:06:48.351)=167
                      [e]Data feed time difference too big abs(14:04:01.583 - 14:06:48.422)=167
                      [e]Data feed time difference too big abs(14:04:01.557 - 14:06:48.750)=167
                      [e]Data feed time difference too big abs(14:04:01.541 - 14:06:48.815)=167
                      [e]Data feed time difference too big abs(14:04:01.571 - 14:06:48.818)=167
                      [e]Data feed time difference too big abs(14:04:01.581 - 14:06:48.819)=167
                      [e]Data feed time difference too big abs(14:04:01.532 - 14:06:48.880)=167
                      [e]Data feed time difference too big abs(14:04:01.567 - 14:06:48.884)=167
                      [e]Data feed time difference too big abs(14:04:01.578 - 14:06:49.281)=168
                      [e]Data feed time difference too big abs(14:04:01.547 - 14:06:49.344)=168
                      [e]Data feed time difference too big abs(14:04:01.584 - 14:06:49.481)=168
                      [e]Data feed time difference too big abs(14:04:01.577 - 14:06:49.679)=168
                      [e]Data feed time difference too big abs(14:04:01.535 - 14:06:49.740)=168
                      [e]Data feed time difference too big abs(14:04:01.554 - 14:06:49.743)=168
                      [e]Data feed time difference too big abs(14:04:01.531 - 14:06:49.806)=168
                      [e]Data feed time difference too big abs(14:04:01.532 - 14:06:49.807)=168
                      [e]Data feed time difference too big abs(14:04:01.562 - 14:06:49.875)=168
                      [e]Data feed time difference too big abs(14:04:01.581 - 14:06:49.944)=168
                      [e]Data feed time difference too big abs(14:04:01.537 - 14:06:50.005)=169
                      [e]Data feed time difference too big abs(14:04:01.560 - 14:06:50.272)=169
                      [e]Data feed time difference too big abs(14:04:01.556 - 14:06:50.338)=169
                      [e]Data feed time difference too big abs(14:04:01.570 - 14:06:50.339)=169
                      [e]Data feed time difference too big abs(14:04:01.571 - 14:06:50.539)=169
                      [e]Data feed time difference too big abs(14:04:01.583 - 14:06:50.540)=169
                      [e]Data feed time difference too big abs(14:04:01.587 - 14:06:50.806)=169
                      [e]Data feed time difference too big abs(14:04:01.539 - 14:06:50.868)=169
                      [e]Data feed time difference too big abs(14:04:01.548 - 14:06:50.869)=169
                      [e]Data feed time difference too big abs(14:04:01.566 - 14:06:50.871)=169
                      [e]Data feed time difference too big abs(14:04:01.569 - 14:06:50.871)=169
                      [e]Data feed time difference too big abs(14:04:01.575 - 14:06:50.872)=169
                      [e]Data feed time difference too big abs(14:04:01.544 - 14:06:50.935)=169
                      [e]Data feed time difference too big abs(14:04:01.574 - 14:06:51.004)=170
                      [e]Data feed time difference too big abs(14:04:01.553 - 14:06:51.069)=170
                      [e]Data feed time difference too big abs(14:04:01.565 - 14:06:51.070)=170
                      [e]Data feed time difference too big abs(14:04:01.575 - 14:06:51.071)=170
                      [e]Data feed time difference too big abs(14:04:01.560 - 14:06:51.136)=170
                      [e]Data feed time difference too big abs(14:04:01.587 - 14:06:51.271)=170
                      [e]Data feed time difference too big abs(14:04:01.529 - 14:06:51.333)=170
                      [e]Data feed time difference too big abs(14:04:01.529 - 14:06:51.334)=170
                      [e]Data feed time difference too big abs(14:04:01.556 - 14:06:51.535)=170
                      [e]Data feed time difference too big abs(14:04:01.582 - 14:06:51.737)=170
                      [e]Data feed time difference too big abs(14:04:01.587 - 14:06:51.737)=170
                      [e]Data feed time difference too big abs(14:04:01.546 - 14:06:51.801)=170
                      [e]Data feed time difference too big abs(14:04:01.556 - 14:06:51.802)=170
                      [e]Data feed time difference too big abs(14:04:01.578 - 14:06:51.803)=170
                      [e]Data feed time difference too big abs(14:04:01.553 - 14:06:51.868)=170
                      [e]Data feed time difference too big abs(14:04:01.556 - 14:06:51.868)=170
                      [e]Data feed time difference too big abs(14:04:01.584 - 14:06:51.870)=170
                      [e]Data feed time difference too big abs(14:04:01.585 - 14:06:51.871)=170
                      [e]Data feed time difference too big abs(14:04:01.588 - 14:06:51.871)=170
                      [e]Data feed time difference too big abs(14:04:01.528 - 14:06:51.932)=170
                      [e]Data feed time difference too big abs(14:04:01.551 - 14:06:52.067)=171
                      [e]Data feed time difference too big abs(14:04:01.553 - 14:06:52.267)=171
                      [e]Data feed time difference too big abs(14:04:01.554 - 14:06:52.267)=171
                      Time 14:04 is the time I restarted source feed from RTL_SDR.


                      To describe what exactly I am doing:


                      1. on RPI with RTL dongle, I run this command:
                      Code:
                      nc -l -p 29999 < 1090 | rtl_sdr -f 1090000000 -s 2000000 -g 50 - > 1090
                      where 1090 is unix pipe created using
                      Code:
                      mkfifo 1090
                      2. on another linux box with enough performance for decoding, I run this:
                      Code:
                      nc 192.168.0.16 29999 | ./dump1090 --interactive --net --ifile -
                      Dump1090 shows correct output with no issues in realtime.

                      3. on the same box as 2 I run fr24feed with only --fr24key=xxxxx parameter (other defaults are ok)

                      Again the reason is that I want to lower load on Rpi to have enough spare power for other jobs (reading weather data from sensors) also running here.

                      Comment


                      • Well, well, well.. I did some more testing and the result is that heavy network traffic of raw 1090 MHz data being transferred produces higher load on Rpi than decoding and sending away in decoded form.

                        Thus there is no reason to not decode ADS-B data on Rpi itself.

                        edit.
                        for my reception range - just cca. 20 nm - it produces network traffic of 30 Mbit/s and more.
                        Last edited by marty_oc; 2014-07-07, 14:13.

                        Comment


                        • The time difference is not the offset of your system clock,, but the offset between information received / information decoded columns in the TCP 30003 port output. The original Basestation software used to delay information by 5 minutes and this check is done exactly to prevent that.

                          Please make sure you are running the latest build of the software and confirm "manually" but watching tcp 30003 output that both columns show the same time and date.

                          Comment


                          • Piopawlu, thanks for clearing things out.
                            Seems that the time offset was caused by buffered data in queue - Rpi was obviously unable to feed raw ADS-B data over network fast enough.
                            Does anyone have an idea how much traffic raw ADS-B data generates? I found 5 GB buffer file full after about hour or two of running RTL_SDR at 1.09 GHz. Obviously my 8 GB SD card was filled very soon

                            Comment


                            • New Command Line Argument

                              Today I switched over from feeding from a Windows 8.1 laptop to a Raspberry Pi running dump1090 and the rpi version of fr24feed. While my box is private, I did note that "ps -ef" reveals my sharing key. Is this key considered secret or is it okay that it can be seen? I was going to suggest that fr24feed have an additional optional command line argument added such as "--keyfile [FILENAME]" so that the key can be read from a protected file and not be visible to the world in ps.
                              T-KISP3 feeder covering KFRG and KISP airports on Long Island.

                              Comment


                              • Do you give the world access to an unlocked screen on your windows PC?

                                If you are using the Pi in the way described here http://forum.flightradar24.com/threa...e-to-feed-FR24 - then it does a few things that are insecure.

                                we run the software as root - we could create a user level account and run the software from there (see http://www.debian-administration.org...ding_new_users )
                                we have not changed the root password

                                I admit I haven't bothered much with the security - I see the Pi feeder as a tool to do a specific job not as a multi-function multi-user machine, If I had another task that needs running I would get another Pi (interestingly I have Pi B+ with 4 USB ports on order, delivery tomorrow)

                                it would probably be better in the first instance that we ensured that the world don't get the access on the Pi they require to run 'ps' in the first place, if they do get that level of access, they can probably read the scripts and the keyfile.

                                Comment

                                Working...
                                X