Recently i noticed that my station (No.399) started to return empty observations to the network. What is strange is that i haven’t touched anything and when trying to run rtl_test on my raspberry to check if the rtl dongle is recognised properly and debug the problem, the command executes as expected, but when i press ctrl+c to stop the test, i get a “floating point exception” error (see attachment) This problem only disappears if i completely restart the raspberry. Furthermore, i noticed that when i return an empty observation of any NOAA satellite, the decoder actually creates a black image, hence the network automatically marks the observation as good which is incorrect, since there is not even audio uploaded. Isn’t that a bug?
Please check the
/tmp/.satnogs/data folder and clean it up, if neccessary!
$ df -h should give you some hints if it is filled up.
Sometimes a bunch of
receiving_waterfall_[..] files are leftover from previous observations. I couldn’t dig deeper into that yet.
Hmmm… It seems that the problem appeared again… This time i checked for storage issues with no luck. My raspberry continues to send empty observations to the network and the only abnormal thing i can spot is the “floating point exception” message while exiting the rtl_test command. a reboot fixes the issue, but this isn’t a permanent solution
journalctrl gives me the following output. I don’t know if it is relevant or if it helps, but i see an error
Oh and really sorry for the quality of images. I didn’t know exactly what to look for
Did you ever find the cause of this problem, or a solution/workaround? I seem to have a similar problem with my station.