@arthur I took the liberty to press the code block button (looks like this: </> ) to make the output more readable.
It looks like your /usr/include/linux/types.h file is damaged. This is what the top of that file looks on my system:
pi@raspberrypi:~ $ head /usr/include/linux/types.h
Could you post yours?
I had the same problem.
For some reason I'd connected NRF_CE to Pin D4 instead of D9.
Simply putting #define MY_RF24_CE_PIN 4 made all the problems go away but I lost a couple of days figuring that out. I assumed the radio was connected fine as it was talking to the gateway, it would just not ACK messages.
The gateway would show:
As you also suspect I think the gateway is rebooting at a regular interval. In my experience this can be due to two causes. Either due to a bug that was present in an older core arduino firmware version or due to the gateway doesn't get enough power to handle sending of messages. Sending messages requires more power than receiving messages.
How did you program the gateway, ie what software and what version of software did you use? How do you power the gateway? Do you use the recommended capacitor close to the radio?
See this post for info on arduino core versions with the bug:
It was solved in version 1.6.18.
Looks like we finally have the tools to keep spam out of the forum after upgrade.
This means a little more wait time for new users to get their posts accepted. But I think it's worth it to keep the evil spammers out.
I had the gateway reboot problem since 2.1.1. After initial startup, the gateway would keep rebooting after receiving messages from nodes. After a while it seems to stabilize.
I tried 2.2.0-rc.1 as you suggested and I don't see that problem any more. Thank you!
@Johandelange I didn't look closely enough at the logs, but I think @mfalkvidd may be right. You may need to assign the node IDs manually in your code if you are using MQTT. The 47uf capacitors might be a little high, but may still work fine. On my radio nodes I typically use 4.7uf, but your situation could be different. It does seem to be working for you.
I found the problem. In an unrelated part of the sketch I was doing an analogue read from an input set up as a digital output. This apeared to generate some sort of error state that fouled up the interface between teh Arduino and the RFM69. After removing the offending statements everything works fine.
@nelsonov Great! There is no need to create an issue. If there is anything to discuss, we can discuss in the pull request thread.
Maybe call it something like "Add MY_DEBUGDEVICE to redirect debug output"