New piAware Troubleshooting

Greetings from a noob to the Raspberry Pi and the piAware scene.

I put my kit together, setup the WiFi, plugged in the SDR/ADS-B dongle w/ antenna… and waited… and checked the logs… and waited… and checked the logs…

I did a whole bunch of reading about piaware statuses and such… and mine looks pretty well inline with all of yours…

My output file is another matter… as once the programs get up and running, I constantly get this time-based update as the system has no data to work with, so it resets itself:



05/06/2015 15:18:21 seconds since last message or startup (2700) less than threshold for action (3600), waiting...
05/06/2015 15:18:21 dump1090 is listening for connections on FA-style port 10001
05/06/2015 15:18:51 0 msgs recv'd from dump1090 (0 in last 5m); 0 msgs sent to FlightAware


I currently live in an apartment building with relatively little line of sight to a large piece of sky, but I should be getting something from my receiver (NooElec R820T), right?

Looking for some troubleshooting help… kicks in the right direction… etc.

Many thanks,
Marc

Double check that your antenna is connected. Other than that, check the LIVE FLIGHT TRACKING link [upper left] for flights near your location. It is possible that the part of the sky that your antenna can see is very quiet.

Good luck and keep us posted as to your situation, Tom

Thanks, Tom.

I’ve moved my piAware around in a different window to see if I can get more clear sky…

Rebooted and gave it some time… ran a piaware-status with these results:


dump1090 is running.
faup1090 is not running.
piaware is running.
dump1090 is listening for connections on port 30005.
dump1090 is listening for connections on port 10001.
piaware is connected to port 10001.
piaware is connected to FlightAware.

My output file looks like this, and shows the fact that I had to manually ask fadump1090 for a restart just a couple of minutes ago:


05/06/2015 19:57:23 ****************************************************
05/06/2015 19:57:23 piaware version 1.20 is running, process ID 2091
05/06/2015 19:57:23 your system info is: Linux piaware 3.18.5-v7+ #225 SMP PREEMPT Fri Jan 30 18:53:55 GMT 2015 armv7l GNU/Linux
05/06/2015 19:57:23 connecting to FlightAware eyes.flightaware.com/1200
05/06/2015 19:57:23 got 'couldn't open socket: no such device or address (Temporary failure in name resolution)' to adept server at eyes.flightaware.com/1200, w
ill try again soon...
05/06/2015 19:57:23 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 19:57:23 no ADS-B producer program seen for 0 seconds (or since piaware started), will attempt to start it next check after 360 seconds...
05/06/2015 19:57:23 no ADS-B producer (dump1090, modesmixer, etc) appears to be running or is not listening for connections on port 30005, next check in 5m
05/06/2015 19:58:08 0 msgs recv'd from (not currently connected to an adsb source); 0 msgs sent to FlightAware
05/06/2015 19:58:23 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 19:59:04 connecting to FlightAware 70.42.6.203/1200
05/06/2015 19:59:04 FlightAware server SSL certificate validated
05/06/2015 19:59:04 encrypted session established with FlightAware
05/06/2015 19:59:04 autoUpdate in /etc/piaware is enabled, allowing update
05/06/2015 19:59:04 manualUpdate in /etc/piaware is enabled, allowing update
05/06/2015 19:59:04 logged in to FlightAware as user lovelacm
05/06/2015 19:59:23 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 20:00:04 server is sending alive messages; we will expect them
05/06/2015 20:00:23 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 20:01:24 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 20:02:23 no ADS-B producer program seen for 300 seconds (or since piaware started), will attempt to start it next check after 360 seconds...
05/06/2015 20:02:23 no ADS-B producer (dump1090, modesmixer, etc) appears to be running or is not listening for connections on port 30005, next check in 5m
05/06/2015 20:02:24 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 20:02:54 0 msgs recv'd from (not currently connected to an adsb source) (0 in last 5m); 0 msgs sent to FlightAware
05/06/2015 20:03:24 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 20:04:24 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 20:05:24 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 20:06:25 no ADS-B data program is serving on port 30005, next check in 60s
05/06/2015 20:07:23 seconds since last message or startup (600) less than threshold for action (3600), waiting...
05/06/2015 20:07:23 dump1090 is listening for connections on FA-style port 10001
05/06/2015 20:07:25 ADS-B data program 'dump1090' is listening on port 30005, so far so good
05/06/2015 20:07:25 i see dump1090 serving on port 10001
05/06/2015 20:07:25 connecting to dump1090 on port 10001...
05/06/2015 20:07:25 piaware is connected to dump1090 on port 10001
05/06/2015 20:07:53 0 msgs recv'd from dump1090 (0 in last 5m); 0 msgs sent to FlightAware


There are several good candidates for ADS-B traffic where I’d expect to be able to pick up something… (Thanks for the note about the real-time tracking link, as I didn’t think about checking to see what could be sending messages… duh.)

I know I’m still using the crappy small antenna that came with the receiver dongle, but is there any good gouge on either checking the dongle or the antenna? Dongle has a steady blue LED, and the antenna’s connections are solid… frustrating.

You could try running the rtl_power utility to test for available signals in your area. Here’s how I run it when I’m testing an antenna. Here is a link to a graph I created in Excel from one of my scans. http://tjowen.no-ip.info/spider4a-amp-split.jpg

…keep us posted, Tom

Log into the Pi box and use the following commands.

  1. first you need to kill the dump1090 application

sudo killall dump1090

  1. run the rtl_power utility, this will create a comma-separated-value (CSV) file which can be openned in Excel and graphed, it will scan the frequencies from 400MHz to 1300MHz and will complete in 10 minutes

sudo rtl_power -f 400M:1300M:1M -g 50 -i 100 -e 10m stock-antenna-scan.csv

  1. reboot to restore everything running once again

sudo shutdown -r now

One other thing… I’m assuming that you have the dongle plugged directly into the Raspberry Pi. With this in mind how good is the 5V power supply running the system? I had a few stability issues early on that I was able to clear up by moving the dongle to a powered USB hub.

…Tom

You might get an improvement by putting the aerial on a metal biscuit tin lid.

Thanks to you both, Tom & Rusty.

Rusty - I’m afraid I don’t have a biscuit tin or lid handy, but I’ll keep my eye out. I always seem to prefer Digestives, for some reason…

Tom - I saw your other thread about running 2 clients off a single antenna, so I started monkeying with the rtl_power utility without success. I’ve rechecked my connections (yes, the dongle is directly connected to the pi’s USB, and the power supply is stock that came with the kit: ANU-050200A, 5V, 2000mA output.

I’ll run through your suggested plan of action and pass along words shortly.

Regards,
Marc

Rusty - I just plotted your location on Google Maps… and discovered that you most likely used to pick up my old chariots (MC-130P out of EGUN) while we would fly low level in Area 5 of the UKLFS.

Well butter my butt and call me a biscuit! It looks like it could be working, now… I’d unhooked it all, and when I plugged it back together and set it in the window for the test you suggested, I suddenly received email from FlightAware that they’re receiving data from my piAware…

Here’s the current output file:


05/06/2015 20:19:50 ****************************************************
05/06/2015 20:19:50 piaware version 1.20 is running, process ID 2074
05/06/2015 20:19:50 your system info is: Linux piaware 3.18.5-v7+ #225 SMP PREEMPT Fri Jan 30 18:53:55 GMT 2015 armv7l GNU/Linux
05/06/2015 20:19:50 connecting to FlightAware eyes.flightaware.com/1200
05/06/2015 20:19:50 FlightAware server SSL certificate validated
05/06/2015 20:19:50 encrypted session established with FlightAware
05/06/2015 20:19:50 autoUpdate in /etc/piaware is enabled, allowing update
05/06/2015 20:19:50 manualUpdate in /etc/piaware is enabled, allowing update
05/06/2015 20:19:50 ADS-B data program 'dump1090' is listening on port 30005, so far so good
05/06/2015 20:19:50 i see dump1090 serving on port 10001
05/06/2015 20:19:50 connecting to dump1090 on port 10001...
05/06/2015 20:19:50 piaware is connected to dump1090 on port 10001
05/06/2015 20:19:50 dump1090 is listening for connections on FA-style port 10001
05/06/2015 20:19:50 piaware received a message from the ADS-B source!
05/06/2015 20:19:50 logged in to FlightAware as user lovelacm
05/06/2015 22:57:46 6 msgs recv'd from dump1090; 3 msgs sent to FlightAware
05/06/2015 22:57:46 dump1090 is listening for connections on FA-style port 10001
05/06/2015 22:57:51 piaware has successfully sent several msgs to FlightAware!
05/06/2015 22:58:16 18 msgs recv'd from dump1090 (12 in last 1m); 15 msgs sent to FlightAware
05/06/2015 22:58:39 server is sending alive messages; we will expect them
05/06/2015 23:03:16 76 msgs recv'd from dump1090 (58 in last 5m); 73 msgs sent to FlightAware
05/06/2015 23:07:46 seconds since last message or startup (345) less than threshold for action (3600), waiting...
05/06/2015 23:07:46 dump1090 is listening for connections on FA-style port 10001
05/06/2015 23:08:16 76 msgs recv'd from dump1090 (0 in last 5m); 73 msgs sent to FlightAware

Looks like the smart thing to do would be to just leave it alone… don’t breathe near it… and see how it reacts for a day or so.

Thanks again.

Great news. You’ll be able to watch your FA stats page as time goes by to see how your coverage area fills in. This will give you a very good idea of how your antenna and location are performing. Then you can plan your next move.

Remember, small changes will let you learn what helps and what doesn’t quicker then big changes. :slight_smile:

…Tom

Thanks, Tom.

Looks like that was a fluke… got some nice hits and then the thing died again.

I’m wondering if I have some sort of connection issue with the dongle or if there’s a short in the cable to the antenna…

Try inspecting the MCX connector on the antenna cable and dongle for damage. Also there should be no movement in the MCX socket on the dongle with the antenna connected. If there is the MCX socket on the dongle could be hiding a broken solder joint.

If all looks good, try a different mini-USB power supply on the Raspberry Pi.

…Tom

sigh

Connections look pretty solid. No movement noted in the antenna / dongle connection.

Thoughts on a “new” mini-USB power adapter to pick up? Mildly frustrated in the fact that brand new piece of kit could be the issue…

Do I need more power available to the Pi? (Interestingly enough, when I tried to run that freq sweep on my own, I got an error -4 that seemed to be for power when I read around the Internet for issues)…

Thanks for your help. Sorry to be the whiny new kid on the block. Was nice to actually get something to “work” for a minute…

I went back and read your 1st post in this thread and seen that you added the USB wifi adapter as well as the dongle. The pi has a limited power supply available to the USB ports. So, yes you have a power issue, but I don’t think a new adapter will help. Try connecting the pi to your home network via a Ethernet cord and remove the USB wifi adapter. Or, plug both USB devices into a powered USB hub.
…Tom

Cheers, Tom.
Thanks to your suggestions, I’ve got a new powered USB 3.0 hub enroute.
I’ll give that a shot on Friday, and we’ll see where we go from there.

As expected, Tom, you were all over the power issue.

Once my powered USB hub arrived, I swapped both the SDR dongle and the USB WiFi adapter over to the hub, just to make sure that there should be no power issues coming from the Pi.

She’s been running smoothly since late this morning, and it’s been interesting to finally see the results flowing in.

I’m sure I’ll harass you folks with more questions in the future. Thanks again for the help.

Great news. Now your on your way. I wonder how hooked you’ll become…

Around Christmas I became interested in the Raspberry Pi and Banana Pi micro-computers for media-centre builds. And found that they worked very well for that purpose. But since I stumbled upon the FlightAware Raspberry Pi build page in late February, I’ve sort of been hooked. And have found a whole re-kindling of an electronics/radio hobby that I haven’t been focused on since I graduated from trade school in 1981. Plus the added bonus of getting to play once again with Linux. Another thing that I’ve been interested in for about 20 years.

So far I’ve built several ground plane antenna and just recently a Franklin variety. It’s been very interesting studying the affect of different components and how they add or take away from how many aircraft can be seen.

Anyhow, I’m starting to get long winded, so have a great weekend, Tom.

I think the Pi’s power issues are serious enough that they should be mentioned in the “Build a PiAware” page, if you use the Nooelec DVB-T stick and EDUP USB WiFi dongle then you’re going to have problems running it from a Pi (as I’ve found out over the last day or two). Attaching a USB power monitor to the Pi input and output (USB) ports I get:



Pi input power: 5.20V 550-650mA.
EDUP WiFi: 4.85V 100mA
DVT-T: 4.87V 270mA (!!)


So the USB devices are drawing 370mA out of ports for which the recommended power draw is 100mA each, the WiFi dongle is OK but the DVT-T is pushing it (in fact even the WiFi dongle may be too much, I noticed the problem when both WiFi and ethernet lost network connectivity after a couple of hours, the Pi was still running but it wouldn’t respond to SSH or pings). I’m going to try using one of these, which has the nice property that the micro-USB is passthrough so I can feed it from the UBEC I’m using without having to make up a new cable/adapter plug.

I’d like to echo the suggestion above that the power issue needs a mention in the “Build a PiAware” page.

I had enormous problems with crashes, feeders stopping and the Pi rebooting itself with the Nooelec and EDUP dongles installed. I put them into a powered USB hub that is designed for the Pi and it has been solid since. I don’t think it is safe to put anything into those ports that requires power from the Pi.

This is the hub I bought. You will probably be able to find something similar local to you, it is important that the hub does not supply power back through the connection to the Pi as this can make it unstable. Most cheap hubs don’t control this.

thepihut.com/collections/raspber … spberry-pi

This is actually less critical with the Model 2s, I’m currently feeding mine from a generic (unmodified) powered hub (which was all I had available on short notice) and it works fine. The original Pi versions were more sensitive to this, but the Model 2 has some changes to the hardware in response to the USB-power issues so you can typically get away with a generic hub. Obviously it’s better to get the Pi-specific hub if you’ve got the time and money, but a generic hub will often do at a pinch (like when you’re leaving the country for a month the following day :slight_smile:.

I have two sites running PiAware and one has been running WiFi and the Dongle just fine for almost 100 days, no power issues at all on the Pi 2 with a 2 AMP Power Supply. Saturday I moved my antenna to a higher location and doubled my receive rate and started having issues with the Pi losing connection to the network where it would drop and then looking at the logs it was losing the WiFi and trying to reconnect. It would eventually but it was very unstable, it seems that it wasn’t an issue for me until my message rate went much higher. Just thought I would share as someone might be reading this like I was thinking you all were crazy mine was working fine without a hub. I just pulled the WiFi adapter and went hard wired since I was able to run CAT5 to the new location anyway and the Dongle alone seems to be running just fine connected directly to the Pi now.