RPi v1.2 running FR24feed v 1.0.23-8 should I upgrade?

[DrRobin]

I am running an old RPi v1.2 and FR24feed version 1.0.23-8 with dump1090-mutability v1.15 dev. I know the feeder is a little old but when I used apt-update it said it was up to date. I think the latest feeder is 1.0.25. Should I update to the latest version, perhaps a manual install or is 1.0.23-8 really the last version for my setup?

Will I gain anything by updating?

i tried a general apt-update, but it failed to find a number of files and didn’t complete,I am not sure how to get anywhere with this, short of downloading a new image and copying my key across. I would like to keep my historical graph data.

I am sure I used to participate in MLAT and although it is still turned on my graphs don’t show any MLAT positions so not sure it is actually doing anything.

[wiedehopf]

FR24 doesn't give MLAT results back to the participants.
Which graphs are you talking about?

The Rpi 1 is rather low CPU so ... the more stuff you install the more problematic it is.

Just start with a new sd-card.
If it's graphs1090, there are hints how to copy / backup the graph data: https://github.com/wiedehopf/graphs1090#graphs1090
If you're gonna start fresh, consider going this path: https://github.com/wiedehopf/adsb-wi...ADS-B-receiver

[DrRobin]

Thanks for the quick reply.

I run graphs1090.

I realise RPi 1 is rather slow, but it has been a good performer and RPi 3 or 4 have clock frequencies which are closer to ADS-B frequencies. I have an RPi3 used on another project, perhaps I will try that sometime. I was regularly getting out to 350 miles but since last October I can't get past 260 miles, the voltage presented by the Pi to the Flightaware Pro is a little low, I will be fixing that with a homebrew powered USB cable (power injector).

graphs1090.JPG

I am also thinking noise in the PSU and/or low voltage is causing the range limit on the FA Pro, nothing else has changed and I tested my antenna (4 x 1/2 wave dipoles, phased) last week, it checked out okay on an analyser and gave better results than a J-Pole at the same height. I have a FA external filter and tried that to see if it is cellular interference, if anything the number of planes and messages/sec went down with the filter in place, so don't think it is cellular interference.

I am rebuilding the spare RPi in to a new outdoor enclosure with screening round the Pi and FA Pro, as well as a [Meanwell] DC-DC converter just to be sure.

I have cloned my SD card and have a second RPI, so can mess about running updates etc, to get it working correctly, then copy the graph data (thanks for the link) across when complete.

I think that for the present I am going to stay with Dump1090-mutability as it gives me the visible horizon limit lines, I don't think Dump1090-FA supports these?

Robin

[wiedehopf]

You mean a range outline from heywhatsthat? Of course it's supported: https://github.com/wiedehopf/tar1090...-range-outline

Anyhow i'm suggesting readsb with tar1090

(dump1090-fa is based on dump1090-mutability so it's pretty much superior in every way, well despite maybe some people having opinions on the appearance of the webinterface.
readsb is a fork of dump1090-fa .... it should work on an RPi1 while dump1090-fa currently will use too much CPU)

[DrRobin]

Thanks for all of the advice. I have a spare SD card and now a duplicate hardware system (RPiv1, SDR, etc) so will give your build a try and then compare to my original Dump1090-mutability build and see which works the best, by swapping SD card on an offline system. I will be able to compare either to my online build and check it is getting better or worse.

Once it is correct I will swap my live system over. It will be good to get an up to date software image as mine doesn’t want to update.

I will let you know how I get on.

Robin

[DrRobin]

Hi,

I have installed the latest Pi version together with readsb, tar1090, graphs and heywhatsthat on my RPiV1 test system, it is using a cheap dongle (waiting for a better 2nd dongle to arrive) and internal antenna. Everything seems to work fine, no issues/errors/warnings and some nice on-screen displays. I haven't put my feeder in yet, don't want it interferring with the live system.

One thing I noticed is that for the same gain as my old system the RSSI figures are around 9dB lower for a couple of fixed sites, but I think the noise floor is also much lower so it must just be a difference in scaling. Either way it holds connections for about the same time. I wil need to get it working with the proper RTL-SDR and decent external antenna to really tell.

Robin

[MRPV26]

Hi!

I noticed I have software version 1.0.27-2 installed. Is this one the most recent one? Should I do apt-get fr24feeder or something like that?

[DrRobin]

All going well except I can't get FR24feed to work. I have tried v26-9 and 27-2 but get the same result.

sudo fr24feed-status
[FAIL] FR24 Feeder/Decoder Process ... failed!

And

May 05 10:05:22 raspberrypi systemd[3720]: fr24feed.service: Failed to determine user credentials: No such process
May 05 10:05:22 raspberrypi systemd[3720]: fr24feed.service: Failed at step USER spawning /bin/rm: No such process
May 05 10:05:22 raspberrypi systemd[1]: fr24feed.service: Control process exited, code=exited, status=217/USER
May 05 10:05:22 raspberrypi systemd[1]: fr24feed.service: Failed with result 'exit-code'.
May 05 10:05:22 raspberrypi systemd[1]: Failed to start Flightradar24 Decoder & Feeder.
May 05 10:05:27 raspberrypi systemd[1]: fr24feed.service: Service RestartSec=5s expired, scheduling restart.
May 05 10:05:27 raspberrypi systemd[1]: fr24feed.service: Scheduled restart job, restart counter is at 4.
May 05 10:05:27 raspberrypi systemd[1]: Stopped Flightradar24 Decoder & Feeder.
May 05 10:05:27 raspberrypi systemd[1]: Starting Flightradar24 Decoder & Feeder...

And
systemctl status fr24feed.service
● fr24feed.service - Flightradar24 Decoder & Feeder
Loaded: loaded (/etc/systemd/system/fr24feed.service; enabled; vendor preset: enabled)
Active: activating (auto-restart) (Result: exit-code) since Wed 2021-05-05 10:26:22 BST; 554ms a
Process: 8745 ExecStartPre=/bin/rm -f /dev/shm/decoder.txt (code=exited, status=217/USER)

And
journalctl -xe
--
-- A stop job for unit fr24feed.service has finished.
--
-- The job identifier is 17593 and the job result is done.
May 05 10:27:35 raspberrypi systemd[9031]: fr24feed.service: Failed to determine user credentials:
May 05 10:27:35 raspberrypi systemd[9031]: fr24feed.service: Failed at step USER spawning /bin/rm:
-- Subject: Process /bin/rm could not be executed
-- Defined-By: systemd
-- Support: https://www.debian.org/support
--
-- The process /bin/rm could not be executed and failed.
--
-- The error number returned by this process is ERRNO.
May 05 10:27:35 raspberrypi systemd[1]: Starting Flightradar24 Decoder & Feeder...
-- Subject: A start job for unit fr24feed.service has begun execution
-- Defined-By: systemd
-- Support: https://www.debian.org/support
--
-- A start job for unit fr24feed.service has begun execution.
--
-- The job identifier is 17593.
May 05 10:27:35 raspberrypi systemd[1]: fr24feed.service: Control process exited, code=exited, stat
-- Subject: Unit process exited
-- Defined-By: systemd
-- Support: https://www.debian.org/support
--
-- An ExecStartPre= process belonging to unit fr24feed.service has exited.
--
-- The process' exit code is 'exited' and its exit status is 217.
May 05 10:27:35 raspberrypi systemd[1]: fr24feed.service: Failed with result 'exit-code'.
-- Subject: Unit failed
-- Defined-By: systemd
-- Support: https://www.debian.org/support
--
-- The unit fr24feed.service has entered the 'failed' state with result 'exit-code'.
May 05 10:27:35 raspberrypi systemd[1]: Failed to start Flightradar24 Decoder & Feeder.
-- Subject: A start job for unit fr24feed.service has failed
-- Defined-By: systemd
-- Support: https://www.debian.org/support
--
-- A start job for unit fr24feed.service has finished with a failure.
--
-- The job identifier is 17593 and the job result is failed.


I am pretty sure I followed the instructions here (https://github.com/wiedehopf/adsb-wi...ADS-B-receiver) and nothing went wrong.

Any suggestions?

[wiedehopf]

Yes, check the log as instructed in the guide: https://github.com/wiedehopf/adsb-wi...-fr24-optional


sudo journalctl -u fr24feed --no-pager

Is tar1090 (using reasdb data) working?
That's the local interface.

Also because i see something strange, what's the output for this:

file /bin/rm

[DrRobin]

Hi,

Okay, I found this post (17/50) in https://discussions.flightaware.com/...-fr24/70143/18

Which had the following

I put the following in and it fixed the problem.

sudo wget -O /etc/systemd/system/fr24feed.service https://raw.githubusercontent.com/wi...24feed.service
sudo systemctl enable fr24feed
sudo adduser --system --no-create-home fr24
sudo systemctl restart fr24feed
sudo journalctl -u fr24feed -ef

It seems to be the 'adduser' line which is missing in the guide.

Robin

[wiedehopf]

No it's in the guide: https://github.com/wiedehopf/adsb-wi...-fr24-optional
( and not i haven't changed it just now)

[DrRobin]

To answer your earlier questions,

sudo journalctl -u fr24feed --no-pager
[Robin] I printed a section of this above, it went on for several pages so didn't show it all.

Is tar1090 (using reasdb data) working?
That's the local interface.
[Robin] Yes, tar1090 & readsb are both working correctly. They are the only things installed, dump1090-mut has never been installed on this new build.

Also because i see something strange, what's the output for this:

file /bin/rm
[Robin] I didn't get round to trying this as the following fixed it.

I started again, using

sudo apt update
sudo apt install -y perl wget
sudo apt remove -y fr24feed
sudo adduser --system --no-create-home fr24 sudo rm /tmp/fr24 -rf mkdir /tmp/fr24 cd /tmp
sudo rm /tmp/fr24 -rf
mkdir /tmp/fr24
cd /tmp

wget -O fr24.deb https://repo-feed.flightradar24.com/...27-2_armhf.deb

sudo wget -O /etc/systemd/system/fr24feed.service https://raw.githubusercontent.com/wi...24feed.service
sudo systemctl enable fr24feed
sudo adduser --system --no-create-home fr24
sudo systemctl restart fr24feed
sudo journalctl -u fr24feed -ef

I already had my user settings and it sprang in to life, working correctly. I am not sure if that second 'adduser' line is really required or if something went wrong first time I installed from the start, but it is now fixed.

Thanks for all of your help.

Robin

[DrRobin]

I appear to have a problem with FR24Feed, I am getting much lower positions reported in in FR24 Stats screen since moving from FR24feed v1.0.23 to 1.0.27-2, although the number of hits is broadly the same. In addition if I look in tar1090 and graphs1090 the distance, number of messages and number of positions is about the same before and after moving from dump1090-mutability and v1.0.23 to readsb and v1.0.27.

For most of the last week my hits have peaked at 21,000/hr and 10,000 positions (09:00 UTC), now the hits are still peaking at 21,000 but positions have dropped to peak around 4,000 (09:00 UTC). The stats on FR24 feeder page for positions and peak distance are significantly lower for every hour of the day since upgrading. If I watch tar1090/graphs1090, wait a while and then look on FR24 stats page, a number of aircraft I see just don't appear.

It is almost as if I am just not forwarding a significant amount of positions.

I have tried running sudo service fr24feed status, but it doesn't show any faults.

Is there anything I can do to get more information or is this a known problem?

[DrRobin]

I have removed FR24Feed 1.0.27-2 and gone to 1.0.26-9, same result and then to 1.0.23-8, again same result. I did notice a significant difference in the ADSB Tracks seen graph, see below:

graphs1090-2.JPG
Data up to 4th May was with dump1090-mutability, the 5th was the day of change over (note the gap) and then the second half of the 5th and 6th May was with readsb (I copied the historical data). The tracks with more than one message are roughly the same, but the tracks with single message are much reduced (about a half to a third). Nearly all of the other graphs are roughly the same before and after the SD card change.

Is this a clue as to why I am getting far fewer positions to FR24 with readsb than I was with dump1090-mutability?

[wiedehopf]

Single message track is just noise.

FR24 statistics is not something i speculate about as i can't look under the hood.
Could be they changed the reporting of stats with the versions.
But really if your local stats show you performance is fine then performance is fine.