If you take over 5 minutes for the question, please continue in Discord or a Github Issue.
Q.1 : logger stopping after a couple of hours on 1.14
Summary: 1.14 Cherry picked PR from Beat Keung that restarts the logger whenever reaching max file size from Nuttx. However, it is stopping logging after sitting for long periods of time (~5 hours)
Answer vfat size is confusing. Is there a way to have vfat larger from logging.
SD card that is being used: one from Holybro. One possible solution is to stream the log to a companion computer.
Daniel: 2 solutions
format in linux, but add a bad block check that goes through and does a block check compared to a quick reformat.
make sure that SD card is actually a good one. not just a cheap one
fat filesystem is not great. looking at moving fram to littlefs. using littlefs on SD card as an option could be interesting. This is a proper failsafe filesystem, but you have to know what you are doing to use it.
Q2: RemoteID, and is it going to be released in PX4 at some time?
Answer: Not going for 1.14.1 release but will probably be a point release.
Justifies being included in a minor release. It is a range of different things where different autopilots are implementing it differently. There is a bit of hesitation as there is no
Q3: Working on tethered drone, changed PX4 fork to deal with battery draw, is this something interesting to upstream? The main thing that has been done is ensuring not too many amps are drawn from the tether.
Answer: A battery architecture is going to be most interesting. We should capture a more broad set of battery management as things have been added over time. There might be things in commander that might be too battery specific, and need to be generalized. Best to talk more offline about what battery architecture looks like in PX4
I have an issue where the logger stopping after a couple of hours.
Sadly we are doing long flight, so it’s a big issue for us since we want to test our system for about the same amount of time as the flights and also need some debugging for those.
But alas, it is still present in our testing, where when we leave the system for about 5 hours, we come back and the logger module has stop logging without any error in the dmesg or in the last ULog that was generated.
I ran multiple time tests and 5 hours seems to be the limit which produce a ~900MB ULog file. I’ve just started another test with a freshly formatted SD card, and I’ll see what the outcome is.
Here are a couple of hypotheses of why I think it’s stopping the logging:
There’s no more space in the SD card
Here’s a screenshot of the freshly formatted SD Card, and I only have, 3324M available. I’ll test this by letting it run as long as I can and see if the ULog match the file size of the available memory left.
I’d like to know how can I make this vfat partition bigger?
There’s a limit to the size a file can be in NuttX
Which would correlate with what I’ve experienced, but with the PR it “should” be fixed?
Anyway, if there’s some idea of what I can do to test to confirm where the problem lies that would be greatly appreciated and also if you have any idea on how to fix it that would be stellar also.
Thanks!
Update #1:
So I left it run all night and it now the log is about 1.7Go. So formatting the SD card was important. So I’d like to know if there’s a way to put something like a 32Go SD card and have a vfat that’s really big for 2-3 days flights.
And also I didn’t see any logs in the dmesg regarding that the logger had stop logging or that it had an issue. Maybe it’s because the dmesg doesn’t give you all the information? Like I was only seeing maybe 30 or so lines of log.
Community Announcement
Community Q&A (No deep technical discussions)
Guideline for asking a Question
General Discussions