As an FYI, the unit locks up in my office with full time AC. The electrical circuit is an online UPS and I have business 20MB upload ISP. The cron job would fix the anomalies.
I’ll add a heat sync to help dissipate heat from the chip.
The unit locked up this morning and I remotely POR the unit, the next time this happens I’ll get a screen shot of the status.
Please let me know if you come up with a solution!
Regards,
Pat
Just to clarify, does the web interface lock up or the entire device? How do you remotely reset it?
Only the web interface locks up! The unit is responsive and I can runs commands.
I have a netbooter that I access via the web and it cycles the power with a 15 second delay.
100% agree with that statement which is why I described my solution as ‘cheesy’. 
Are you hard wired to a router or using a WiFi dongle to connect to the internet?
Above you mentioned being able to ping both a public and private IP for your RPi. Would I be correct in assuming that this means you’ve made the dump1090 web interface publicly available? This is why I ask this. Basically, dump1090 will abruptly die if a bad URL is requested.
When I built my box at home, I initially had the web interface publicly available as well, mostly cause it looked cool and I thought my friends and family might enjoy checking it out from time to time. But a couple days after getting it set up, it would stop feeding positions to FA and simultaneously the web interface would stop responding. Investigating via top/ps showed that dump1090 was dying in these cases. So I disabled the port forwarding that made the interface publicly available, and I have had no failures since then. (this was roughly a week ago)
There are also some possible security issues stemming from exposing the web interface. Overall, it’s not recommended to have the web interface open to the public. If I’m reading your comments correctly, then I’d advise that you disable public access and then see if your problems go away.
I liked having access via the web while I was out of the office, I’ll take your advice and diable port forwarding.
Thx,
Pat
Hey everyone, we have listened and responded… PiAware 1.15 has automation to detect dump1090 not sending and will attempt to restart it.
If your system has a dump1090 init script in /etc/init.d, in particular fadump1090.sh but anything with the string “dump1090” in it will work, then if an hour goes by and dump1090 has not forwarded any messages, piaware will invoke the dump1090 script with a “restart” argument.
If you have the FlightAware version of dump1090 installed on your machine with the full-install option, then after upgrading that to 1.15 and installing, the “stop” and “restart” arguments to the init script will work properly. This is vital. (We will have a dump1090 Debian package soon.)
It’s available
as a Debian install package for the Raspberry Pi (upgrade instructions at flightaware.com/adsb/piaware/upgrade)
as an image you can copy to an SD card flightaware.com/adsb/piaware/build)
via source code on the github repos at FlightAware · GitHub.
The reason piaware waits an hour before restarting dump1090 is so that sites that don’t get a lot of traffic won’t be triggering restarts of dump1090 for no reason. We may revisit this or create a more flexible adaptation.
Thanks for participating and thanks for your patience. I think these changes will really help with the hanging-dump1090 problem.
Thank you for helping us help you!
I’ll perform the upgrade and report back with my findings!
The upgrade just worked out fine, thx!
Upgrade successful!
The unit has been frozen for a few hours and the restart hasn’t restored the operation (that is if the restart ran?).
Is there anything you would like for me to gather in log files, etc. to help diagnose the problem?
I didn’t get a chance to stop the port forwarding this weekend but I’ll stop it this morning.
Did the entire device freeze or just the web interface? Are you able to ping it? Can you access the /tmp/piaware.out file?
The unit is alive and I can putty into it.
I have attached the piaware.out for your review;
pi@piaware ~ $ less /tmp/piaware.out
10/20/2014 08:40:58 49977 msgs recv’d from dump1090 (5 in last 5m); 50461 msgs sent to FlightAware
10/20/2014 08:45:58 49988 msgs recv’d from dump1090 (11 in last 5m); 50473 msgs sent to FlightAware
10/20/2014 08:50:58 49998 msgs recv’d from dump1090 (10 in last 5m); 50484 msgs sent to FlightAware
10/20/2014 08:55:58 50000 msgs recv’d from dump1090 (2 in last 5m); 50487 msgs sent to FlightAware
10/20/2014 09:00:58 50017 msgs recv’d from dump1090 (17 in last 5m); 50505 msgs sent to FlightAware
10/20/2014 09:05:58 50069 msgs recv’d from dump1090 (52 in last 5m); 50558 msgs sent to FlightAware
10/20/2014 09:10:58 50097 msgs recv’d from dump1090 (28 in last 5m); 50587 msgs sent to FlightAware
10/20/2014 09:15:58 50106 msgs recv’d from dump1090 (9 in last 5m); 50597 msgs sent to FlightAware
10/20/2014 09:20:58 50117 msgs recv’d from dump1090 (11 in last 5m); 50609 msgs sent to FlightAware
10/20/2014 09:25:28 dump1090 is listening for connections on FA-style port 10001
10/20/2014 09:25:58 50117 msgs recv’d from dump1090 (0 in last 5m); 50610 msgs sent to FlightAware
10/20/2014 09:30:28 dump1090 is listening for connections on FA-style port 10001
10/20/2014 09:30:58 50117 msgs recv’d from dump1090 (0 in last 5m); 50611 msgs sent to FlightAware
10/20/2014 09:35:27 dump1090 is listening for connections on FA-style port 10001
10/20/2014 09:35:58 50117 msgs recv’d from dump1090 (0 in last 5m); 50612 msgs sent to FlightAware
10/20/2014 09:40:28 dump1090 is listening for connections on FA-style port 10001
10/20/2014 09:40:58 50119 msgs recv’d from dump1090 (2 in last 5m); 50615 msgs sent to FlightAware
10/20/2014 09:45:27 dump1090 is listening for connections on FA-style port 10001
10/20/2014 09:45:58 50120 msgs recv’d from dump1090 (1 in last 5m); 50617 msgs sent to FlightAware
10/20/2014 09:50:58 50171 msgs recv’d from dump1090 (51 in last 5m); 50669 msgs sent to FlightAware
10/20/2014 09:55:58 50216 msgs recv’d from dump1090 (45 in last 5m); 50715 msgs sent to FlightAware
10/20/2014 10:00:58 50226 msgs recv’d from dump1090 (10 in last 5m); 50726 msgs sent to FlightAware
10/20/2014 10:05:58 50228 msgs recv’d from dump1090 (2 in last 5m); 50729 msgs sent to FlightAware
10/20/2014 10:06:12 lost connection to faup1090, disconnecting…
10/20/2014 10:06:17 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 10:07:17 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 10:08:17 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 10:09:17 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 10:10:17 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 10:10:27 no ads-b producer (dump1090, modesmixer, etc) appears to be running or is not listening for connections on port 30005, next check in 5m
10/20/2014 10:10:57 50228 msgs recv’d from dump1090 (0 in last 5m); 50729 msgs sent to FlightAware
10/20/2014 10:11:17 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:25:27 no ads-b producer (dump1090, modesmixer, etc) appears to be running or is not listening for connections on port 30005, next check in 5m
10/20/2014 13:25:57 50228 msgs recv’d from dump1090 (0 in last 5m); 50768 msgs sent to FlightAware
10/20/2014 13:26:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:27:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:28:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:29:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:30:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:30:27 no ads-b producer (dump1090, modesmixer, etc) appears to be running or is not listening for connections on port 30005, next check in 5m
10/20/2014 13:30:57 50228 msgs recv’d from dump1090 (0 in last 5m); 50769 msgs sent to FlightAware
10/20/2014 13:31:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:32:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:33:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:34:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:35:26 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:35:27 no ads-b producer (dump1090, modesmixer, etc) appears to be running or is not listening for connections on port 30005, next check in 5m
10/20/2014 13:35:57 50228 msgs recv’d from dump1090 (0 in last 5m); 50770 msgs sent to FlightAware
10/20/2014 13:36:27 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:37:27 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:38:27 no ADS-B data program is serving on port 30005, next check in 60s
10/20/2014 13:39:27 no ADS-B data program is serving on port 30005, next check in 60s
As you can see it stopped responding at 1006, I’ve included the end of the file for reference.
I have installed an external antenna and looked at the log files, it appears that the cron didn’t restart the fadump1090.sh batch file. I manually did it and it came back up and started running. The port forwarding has been canceled, I’ll let you know if any other problems come up.