nrf_cli w/o serial connection - reset loop

[pctj101]

Using FreeRTOS to run nrf_cli. When Serial is not connected (as in RX,TX,GND) the board goes into a reset loop.

There are a few things that seem to help:

1. Don't run cli_start()
2. Don't run the thread that handles nrf_cli
3. Tie NRF52 RX to Vcc 3.3V

nrf_cli_process doesn't seem to have anything to do with it as NOT calling it after cli_start still has boot loops.

Has anyone seen anything like this before?

[Jørgen Holmefjord]

Have you tried debugging, to see if any of the function calls return non-zero error codes? The default behavior when an error is detected is to reset the board.

[pctj101]

Thanks for the informative link. I'll give it a shot. I'm using gcc+daplink on OSX so it's really not cooperating with debugging (until I learn more). Perhaps I can alter things a little bit and have similar info logged to serial or something like that.

It's helpful to see that the normal error handling is a reset. Thanks! I would suppose that it's probably simply RX "not looking right" causing the board to reset.

With a little more playing I noticed that it's not a boot only problem. Disconnecting the serial cables then touching the Rx/Tx pins causes a reset.

If it turns out that RX is simply "not right" which is likely the case, is there a formal way to "ignore" the error?

[Jørgen Holmefjord]

You can check the error code manually instead of passing it to APP_ERROR_CHECK(). If you don't care about the specific error, just ignore it. You can still pass critical errors to APP_ERROR_CHECK() afterwards-