Problem detecting an RNodeInterface on certain computers. What's happening?

[wisepythagoras]

I'm running int an issue getting two Heltec LoRa32 v2 boards on separate computers to talk to each other. They both have version 1.79 of the firmware installed. On my main computer I'm able to get Reticulum (and subsequently Nomadnet) to connect to it (via /dev/ttyUSB0), but on any other computer I try Reticulum doesn't detect the device. I switched them around to make sure one of them is not broken, but they both work fine (I checked with rnodeconf -i /dev/ttyUSBX as well).

Both computers run on the same configuration and the same version of RNS, LXMF, and NomadNet.

This is the output I get:

[Error] Reticulum will attempt to reconnect the interface periodically.
[Notice] Serial port /dev/ttyUSB1 is now open
[Error] Could not open serial port for interface RNodeInterface[RNode LoRa Interface]
[Error] The contained exception was: Attempting to use a port that is not open
[Error] Reticulum will attempt to bring up this interface periodically

After the above error it loads the TUI and then runs the readLoop but it's never able to connect, so it throws me to the branch below:

Reticulum/RNS/Interfaces/RNodeInterface.py

Lines 385 to 387 in 38dd63a

	if not self.detected:
	RNS.log("Could not detect device for "+str(self), RNS.LOG_ERROR)
	self.serial.close()

My configuration is basically a copy-paste of the RNodeInterface config in the docs. The only thing I changed is the frequency.

On the machine(s) that Reticulum doesn't connect to a working device I hardcoded the following and it worked:

        # self.serial.close() 
        self.display = True

But I was curious as to why the device was not detected. I saw the following code in readLoop, but after debugging I noticed that this code never runs.

Reticulum/RNS/Interfaces/RNodeInterface.py

Lines 988 to 992 in 38dd63a

	elif (command == KISS.CMD_DETECT):
	if byte == KISS.DETECT_RESP:
	self.detected = True
	else:
	self.detected = False

So I printed the raw command and byte combos to figure out what happens (See below). The command 0x45 is not defined in RNodeInterface.KISS, so I'm at a loss. What's happening here?

[2025-01-05 18:22:54] [Notice] Opening serial port /dev/ttyUSB0...
Cmd: 0x25 Byte: 0x25
Cmd: 0x25 Byte: 0x1
Cmd: 0x25 Byte: 0x92
Cmd: 0x25 Byte: 0x0
Cmd: 0x25 Byte: 0x1
Cmd: 0x25 Byte: 0x1
Cmd: 0x25 Byte: 0x92
Cmd: 0x25 Byte: 0x0
Cmd: 0x25 Byte: 0x1
Cmd: 0x45 Byte: 0x45
Cmd: 0x45 Byte: 0x20
Cmd: 0x45 Byte: 0x28
Cmd: 0x45 Byte: 0x39
Cmd: 0x45 Byte: 0x36
Cmd: 0x45 Byte: 0x37
Cmd: 0x45 Byte: 0x36
Cmd: 0x45 Byte: 0x30
Cmd: 0x45 Byte: 0x31
Cmd: 0x45 Byte: 0x35
[...]
Cmd: 0x45 Byte: 0x72
Cmd: 0x45 Byte: 0x6f
Cmd: 0x45 Byte: 0x72
Cmd: 0x45 Byte: 0xd
Cmd: 0x45 Byte: 0xa
[2025-01-05 18:22:54] [Error] Could not detect device for RNodeInterface[RNode LoRa Interface]
[2025-01-05 18:22:54] [Notice] Serial port /dev/ttyUSB0 is now open
Cmd: 0x45 Byte: 0x45
Cmd: 0x45 Byte: 0x20
Cmd: 0x45 Byte: 0x28
Cmd: 0x45 Byte: 0x39
Cmd: 0x45 Byte: 0x36
[...]
Cmd: 0x45 Byte: 0x28
Cmd: 0x45 Byte: 0x32
Cmd: 0x45 Byte: 0x32
Cmd: 0x45 Byte: 0x37
Cmd: 0x45 Byte: 0x29
Cmd: 0x45 Byte: 0x3a
Cmd: 0x45 Byte: 0x20
[2025-01-05 18:22:54] [Notice] RNodeInterface[RNode LoRa Interface] is configured and powered up
Cmd: 0x45 Byte: 0x47
Cmd: 0x45 Byte: 0x50
Cmd: 0x45 Byte: 0x49
Cmd: 0x45 Byte: 0x4f
Cmd: 0x45 Byte: 0x20
Cmd: 0x45 Byte: 0x6f

[markqvist]

It could be that something else is trying to access the port on that computer. Some of those hex sequences spell out stuff like ror\r\n (which could be the ending of "error") and (2): . My best guess is that the firmware is crashing for some reason, or that something else is sending errant data on the serial device.

Can you check that you don't have anything else that is trying to access the port?

[wisepythagoras]

I checked last night. I unplugged the device so it wasn't on /dev/ttyUSB0, I plugged in something else (with a serial port connector and an LED) in its place and plugged the LoRa in a different USB port so it was on /dev/ttyUSB1 now. The one on /dev/ttyUSB0 wasn't accessed by anything else (the LED didn't light up). The LoRa board still returned the same commands and bytes. Also, I tried the same on my Raspberry Pis, but had the same result. The workaround in my original message works, but it doesn't solve the issue

I looked at the received bytes again though, because your comment made me wonder. I took all the bytes and ran them through this snippet:

print(''.join([chr(x) for x in errs]))

And this is what I got:

E (9676015) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676018) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676021) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676024) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676028) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676033) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676039) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676045) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676051) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676057) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676062) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676068) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676074) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676080) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676086) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676092) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676097) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9676103) gpio: gpio_set_level(227): GPIO output gpio_num error
E (967_1num error
E (9677748) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9677754) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9677760) gpio: gpio_set_level(227): GPIO output gpio_num error
E (9677766) gpio: gpio_set_level(227): GPIO output gpio_num error

I don't get this on my main computer though, which is weird, because I'm running the same versions everywhere. And I have nothing connected to the GPIO pins