Gateway only receives first message

1

Hello,
we decided to start using Chirpstack. We bought the RAK2245 and we setup the CS server remotely from the Gateway. Our sensor is setup for US freq. so we have setup the GW for US channels.
The transmission from the sensor is received as we can see the “UnconfirmedDataUp” (and down) message in the app. server.
The problem is that the second (or any thereafter) message transmitted by the sensor are not being received. Not until we reset the sensor, another message is received in the app. server.

I checked the log in the GW and I noticed that although a message is picked up by the GW, it is not doing anything with this… :

Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # RF packets received by concentrator: 1
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # CRC_OK: 100.00%, CRC_FAIL: 0.00%, NO_CRC: 0.00%
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # RF packets forwarded: 1 (28 bytes)
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # PUSH_DATA datagrams sent: 2 (326 bytes)
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # PUSH_DATA acknowledged: 0.00%
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: ### [DOWNSTREAM] ###
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # PULL_DATA sent: 3 (100.00% acknowledged)
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # PULL_RESP(onse) datagrams received: 1 (202 bytes)
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # RF packets sent to concentrator: 1 (22 bytes)
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # TX errors: 0
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # TX rejected (collision packet): 0.00% (req:10, rej:0)

VS

[UPSTREAM]

Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # RF packets received by concentrator: 0
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # CRC_OK: 0.00%, CRC_FAIL: 0.00%, NO_CRC: 0.00%
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # RF packets forwarded: 0 (0 bytes)
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # PUSH_DATA datagrams sent: 1 (111 bytes)
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # PUSH_DATA acknowledged: 0.00%
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: ### [DOWNSTREAM] ###
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # PULL_DATA sent: 3 (100.00% acknowledged)
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # PULL_RESP(onse) datagrams received: 0 (0 bytes)
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # RF packets sent to concentrator: 0 (0 bytes)
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # TX errors: 0

Is this a frequency issue…?
I only setup the freq. plan in the GW (gateway-config) to US. Is there anything else that needs to be setup in the Gateway
Thanks
Henk

2

It’s not clear what you mean. I see no log message of a discarded packet.

The summary messages refer to what has happened in the previous small unit of time (likely 30 seconds or a minute) but your log captures show only the summaries, whatever log messages were generated by the events themselves aren’t in your clipped posting.

Is this a frequency issue…?
I only setup the freq. plan in the GW (gateway-config) to US. Is there anything else that needs to be setup in the Gateway

Quite possibly. There are several distinct sub-bands to chose from for US915. It would appear that your gateway configuration and your node defaults are set to the same sub-band, but if Chirpstack is set to a different subband it might be using a MAC command to tell the node to go to a different set of channels than the gateway supports.

Of course that is assuming that the node is still transmitting at all. Do you have serial debug output from the node that shows that is transmits more packets? Does that include any useful details such as frequency and SF?

3

Hello
I based my question on 2 lines in the logfiles:

The first one (the one we receive)
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # RF packets received by concentrator: 1
Nov 30 21:32:43 rak-gateway ttn-gateway[1144]: # CRC_OK: 100.00%, CRC_FAIL: 0.00%, NO_CRC: 0.00%

and the other we don’t receive:
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # RF packets received by concentrator: 0
Nov 30 21:34:15 rak-gateway ttn-gateway[1144]: # CRC_OK: 0.00%, CRC_FAIL: 0.00%, NO_CRC: 0.00%

When we start the Lora Sensor, we see the message coming in on freq 904.100 (US Band 2 CH 8 - 15))
image

Once that has been received, it seems that chirpstack is asking the node to change the frequency to band 1 (CH 0 - 7) as all the other message transmitted are in band 1. When we put the node in debug mode we can clearly see this happening…

So where can we change the subband in CS?

Thanks

4

It’s been a long time since I’ve dealt with that so perhaps someone will have a better answer but a quick look at documentation suggests you want to see enabled_uplink_channels at Configuration - ChirpStack open-source LoRaWAN<sup>®</sup> Network Server