ValueError: invalid join data rate 3 Lopy LORAWAN

[seb]

Hi,

I have managed to replicate your error, I am now investigating the cause and will get back to you when I know more.

[rubencg195]

Hi @seb ,

Sure I am going to send them to your email. Thank you very much.

[rubencg195]

@seb the link is https://docs.pycom.io/chapter/tutorials/lora/lorawan-nano-gateway.html

[seb]

Hi,

I located the cause for your issue, when we updated our LoRa stack to allow users to change regions we added this: https://github.com/pycom/pycom-micropython-sigfox/blob/master/esp32/mods/modlora.c#L1664

This was based on a recommendation in the LoRaWAN spec and I believe it will be relaxed in a future firmware update. For now just make sure when you run lora.join, you set the dr to either 0 or 4.

[rubencg195]

Hi @seb, the only problem is that when I put 0 gives the same error and with 4 it hangs out.

[seb]

which line of code are you changing?

[rubencg195]

@seb line 32 where it says lora.add_channel

[seb]

@rubencg195

You need to adjust the lora.join function not the add_channel. The join has to be DR 0 or 4

[Barillaro]

Hello,

I am having similar issue with same example published on https://github.com/pycom/pycom-libraries/blob/master/examples/lorawan-nano-gateway/otaa_node_US915.py
I have modified line 32 at the end like these:
dr=0 (hang)
dr=1 File "<stdin>", line 32, in <module> ValueError: invalid join data rate 1
dr=2 File "<stdin>", line 32, in <module> ValueError: invalid join data rate 2
dr=3 File "<stdin>", line 32, in <module> ValueError: invalid join data rate 3
dr=4 (hang)
dr=5 File "<stdin>", line 32, in <module> ValueError: invalid join data rate 5
dr=6 File "<stdin>", line 32, in <module> ValueError: invalid join data rate 6
dr=7 File "<stdin>", line 32, in <module> ValueError: invalid join data rate 7
dr=8 File "<stdin>", line 32, in <module> ValueError: invalid join data rate 8

[jclo13]

Hello @seb,

I have the same problem, I change the DR to 4 or 0 and the Lopy4 hangs out. Do you have a solution?

thank you

[jmarcelino]

@jclo13
This is a bug expected to be fixed in the next firmware release. Sorry for the inconvenience.

If you are able to compile a custom firmware there is a fix already on master (compile instructions are there too)

https://github.com/pycom/pycom-micropython-sigfox

[Tombee]

@jmarcelino When do you anticipate the next f/w release?

[jmarcelino]

@tombee
I'm not sure sorry, maybe next week

[jclo13]

@jmarcelino do you have news about the new firmware release?

thank you in advance.

Jose

[jclo13]

I tried whit the last Pybytes firmware and it works, but the memory space is less. And it needs internet to function, right?

[Burgeh]

Any update on how this works with the stable firmware release? (1.17.3.b1?)

[jclo13]

@Burgeh, the firmware 1.18 is already published,

@jmarcelino, I updated the firmware of my lopy4, the problem mentioned before disappeared, but I have another problem, the lopy4 connect successfully whit TTN but it lost connection after the deepsleep, in other words, the lopy4 goes to deep sleep, it wakes up and it think that is connected, it send the data, the gateway receive the data, the TTN Gateway Traffic window shows the packet arrived, but the APPLICATION DATA window doesn´t show anything, if I made the connection procedure again, it shows the data again. Can you help me?, please

thank you

Jose.

[Xykon]

@jclo13 Are you running lora.nvram_save() before going to deepsleep and lora.nvram_restore() after waking up?

[robert-hh]

@jclo13 Deepsleep wakeup is a reset. So before going to deepsleep, you have to save the LORA state tio nvram and restore it after that from nvram. No need to run join again, because the credentials are still valid. Actually, once a device has being joind and the credentials are established, there is no need to join with again, at least with OTAA.

The problem you see may be caused by replaying packet numbers. You can verify that in the TTN console by unchecking the box "frame counter checks" in the devices settings. For testing this is an acceptable setting.

[jclo13]

@Xykon yes, I am using lora.nvram_save() and then lora.nvram_restore().

@robert-hh, I unchecked the box mentioned and it worked, thank you. but you mention for testing, how can I solve the problem if I want to check the counter frame?

Jose