Yveaux

Can you prepare a PR?

If it is using an nRF24L01 it should be possible to sniff the packets. Are you sure other parameters, like bitrate, are set correctly?

As discussed on the forum (https://forum.mysensors.org/topic/11462/mqtt-gw-with-rfm69-on-rpi), very likely related to signing, as it is reported to occur on different gateways, with different radios, when using signing.

Found the issue: Message signing requests a nonce using the _process() method, but this is called from within the same _process() method, hence the 1-level deep recursive call reported by...

It appears to happen in more cases though, judging from logs on the forum: Requesting nonce, log from gw: ``` 47476 SGN:SKP:MSG CMD=3,TYPE=16 47482 TSF:MSG:SEND,0-0-123-123,s=255,c=3,t=16,pt=0,l=0,sg=0,ft=0,st=OK: 47491 SGN:SGN:NCE REQ,TO=123 47493 !MCO:PRO:RC=1...

As far as I know mosquitto has no option to batch delete retained messages. As a workaround you can publish empty messages to retained topics to clear them.

@virtual-maker hey man, you beat me to it! 👍 I took a quick look and this is basically identical to what I did. On the other hand, I'm not happy...

There are many more places in the nRF24 driver that make use the CE-pin, e.g the RF24_sleep() function that you now call instead. This patch does not seem complete.

But the CE pin has a function in the nRF24 communication protocol. Will the nRF24 still function correctly **in all cases** when tied to VCC?

Looking at the nRF24L01+ datasheet ("nRF24L01+ Single Chip 2.4GHz Transceiver Product Specification v1.0") it contains a number of references to toggling the CE pin, or setting it low: 6.1.5 TX...