Troubleshooting

I've found that: char *protocolMyMessage2Serial(const MyMessage &message) { (void)snprintf_P(_fmtBuffer, (uint8_t)MY_GATEWAY_MAX_SEND_LENGTH, PSTR("%" PRIu8 ";%" PRIu8 ";%" PRIu8 ";%" PRIu8 ";%" PRIu8 ";%s\n"), message.getSender(), message.getSensor(), message.getCommand(), message.isEcho(), message.getType(), message.getString(_convBuffer)); return _fmtBuffer; } We put in the print "message.getEcho()", but this isn't echoed message. I think we should put "message.getRequestEcho()" here.

@TheoL there are two pull requests adding support for newer versions. One is guaranteed to break again. One should not break in the future, but will instead break all existing sketches. I’m sure a third option would be welcome if you write the code and submit.

I have mixed Nanos, Pro Minis, ESP8266 and Unos and no problem. I do however have mixed NRF24L01 and NRF24L01+ due to a bad AliExpress saler that stated he sold NRF24L01+ which in fact were not the + version. So I have set my baudrate on all nodes to 1mb. In order to be helpful we need the log on the Uno.

No, it does not recover. The Problem is not easy to debug. I now want to build a serial gateway. With that, I am more flexible to test. Maybe to run MySensors on a second raspberry pi with a standard Home Assistant HASS-Os image. Or I check the MySensors network with MyController. At this moment i run the whole system on 64 Bit raspbian, with HA in docker, which is a little bit experimental and i had to fix some lines in the ethernet gateway to compile it for 64 bit. So this is experimental, too.

Hi OSD, thanks for your input. Those are the 4 relays on IO-MCU. I buzzed out the circuit and found that I had an open circuit on the 12v line to the relays. Furthermore 2 of the relays indicator LEDs were illuminating, 2 were not. I suspect, when I first juiced up the circuit on this MCU, the 2 relays fired, but I'd made the trace too thin for the required current. I wired up some jumper wire on the reverse of the PCB and the 2 relays started working. The 2 that remain inoperable? https://forum.arduino.cc/t/using-pins-a6-a7-on-pro-mini-compatible/118050 A6 and A7 can only be analogue inputs, cannot be digitally written to..... - so this part of my PCB is now redundant. However, all in all, some good fault finding, found the two issues, and I'm now going away to re-think my PCB trace widths and reading the datasheets more accurately ;) Thanks

@ctodor You are very welcome

Hm, maybe the problem is that your rPI has a short interruption in the WIFI connection, the MQTT TCP connection is interrupted and does not autostart again. Maybe there is something in the logs. Maybe you can find something about reconnects in one of these: journalctl --unit=systemd-networkd journalctl --unit=wpa_supplicant You could try to run wpa_cli in daemon mode to react to disconnects and connects. Create a script like this: #!/bin/bash case "$2" in CONNECTED) <your command to start the gateway>; ;; DISCONNECTED) <your command to stop the gateway>; ;; esac then, start wpa_cli in deamon mode: wpa_cli -a /path/to/your/script #use sudo if your gateway commands need sudo If that helps, you could create a service from this command. Regards, Edi

Don't ask me why, but now it works. I changed nothing! Went to my girl over the weekend, coming back, the node was registred. Maybe a jammer went away?

@Mizar03 sorry for the late response. getStream() only works for firmware transfers. I'd suggest getString(), this should also work for all types of payloads, as they are normaly ASCII and not binary data.

I don't know if any of you guys are still around and using this stuff - I moved and am reviving a few of these 51/52 boards that I had built. Prior to moving I had decent luck with all these fixes though I don't quite remember where I was with any of this stuff seeing as it was 3 years ago. I had a few motion sensors and some temp/humd based on these fixes and they were working great, no dips. Not sure any of you will ever see this but hey it all worked.

Solved. Issue with power supply. Added a capacidade to power source.

Hi all, Should anyone face the same problem, I found the root cause: brownout threshold. I burned a new bootloader (Optiboot 8.0) without such trigger and the node has been working with used batteries (~2.7V) since May. Hope this helps.

Thanks for the quick response. Yep! When I updated the IDE I forgot to modify the settings (915 MHz) in the new myconfig.h file. Doh! All good now.

@Matt Yes, the landscape there is incredible. Unfortunately, I mostly have to work when I'm there :) Where are you from?