I have on my ADS-B station two Raspberry Pi: the first one (old B + v1.2) running dump1090-fa, PiAware and graphs1090, and the second one (the new model 4) is my Maps / Tracks server, running Virtual Radar Server with Mono. I want to transfer the graphs1090 from the old RPi to the new, without losing the history. Is it possible to do a new installation on this new RPi and configure it to catch the data from the old one? Is it possible to move the history (+ - 6 months) to from the old to the new RPi?
@wiedehopf, thanks for the answer. Yes, I could use the RPi 4 as the SDR, but I just to “move” the graphs1090 from the old RPi to the new, with the history (yes, I saw the instructions on the GitHub). The main doubt is: if I install graphs1090 on the new RPi, is possible to configure it to read the data generated on the old RPi (where in fact dump1090-fa is running)? Graphs1090 can read data from the remote computer?
Yes you can change collectd.conf but it will only read the SDR statistics.
CPU and temperature statistics and stuff like that will be checked on the Pi4.
Warming up this old thread instead of creating a new one…
I am using a spare Raspberry for doing nice graphs for the data of the Airsquitter.
For this i’ve installed readsb in net-only mode collecting the data from the network port.
On top of it Graph1090 now shows the data as it should be.
I don’t need the graphs regarding disk, cpu, etc for that Raspberry, how can these be removed?
Is it that simple by disabling it in the collectd conf or will the graphs remain there but empty?
How do i remove the ADS-B CPU usage which is around 1% and also not necessary?
I don’t really want to complicate the code to make that possible.
If you want to change the code for your local instance, go right ahead.
On the other hand … the “other graphs” don’t really hurt on typical hardware …
Just downloaded graphs1090. Very nice ! I’m curious as to the units on the ADS-B Signal Level where the legend on the vertical axis is “dBFS”. Is that dB relative to the largest signal ('F’ull 'S’trength) seen ? Is there a way to cover this to absolute dBm of the receiver (with the understanding that the dongles not a bench test equipment device). Even a guesstimate of what absolute dBm corresponds to 0 dBFS would be interesting.
It also looks like the Noise is the noise floor? would be nice to see a different color there, you have to look closely to discern the two different flavors of blue and green lines on the graph.
Note being critical, just user feedback. I’ve done a few “lets plot a bunch of IoT data we’ve seen” projects. graphs1090 is very impressive !
I have a PCB showing up in a week or two that is a Low Noise Pre-Amp, and am thinking about how to quantify its impact on the system. I can sweep it and measure gain with my spectrum analyzer, but don’t have a vector analyzer or noise meter. I do have a return loss bridge, so can get the input S11. As with any LNA, gain is easy and in some respects it “just comes with scenery.” Measuring the improvement in the S/N ratio is the hard part.
It looks like between graphs1090 and the …/adsb/status/user/ web page, the info is there to see the impact of the LNA. There’s bench measurements and there is real world observable system performance.
At any rate, thank you very much wiedehopf for all your hard work
Not that i’m aware of … you’d need to get probably empirical data using a test signal at various dBm levels for a certain gain you use on your specific SDR.
Then you’d need to set up a conversion somewhere in either the graphs1090 or dump1090-fa code.
Have fun with that
Thanks, you can change colors by redefining them in /etc/default/graphs1090 … i suppose this needs some documentation
That’s from graphs1090.sh … you can redefine them in the config.
Be aware noise is just the average power for the time no messages are detected.
That means it includes undetected or messages that failed to decode as such it’s highly variable with band contention and overlapping messages which raise the noise.
All the other data you see is actually statistics on aircraft and their signal levels.
(while noise is a time average of the power of everything minus messages)
Thanks for the reply, I’ll move over to the Graphs for dump1090 thread and reply. There is value in keeping the conversations organized, I was just poking around learning what I could.
“Graphs for dump1090” is a huge/long thread! And I thought “Do I need a Filter” was long… Nothing like being a full time developer and support tech for free software, I’ve been in that situation for 10 years now. The challenges of a complex system with a rich “front of the glass” experience dealing with the complex “behind the glass” realities.
I’m curious as to the units on the ADS-B Signal Level where the legend on the vertical axis is “dBFS”. Is that dB relative to the largest signal ('F’ull 'S’trength) seen ? Is there a way to cover this to absolute dBm of the receiver (with the understanding that the dongles not a bench test equipment device). Even a guesstimate of what absolute dBm corresponds to 0 dBFS would be interesting.
