How to reset chirpstack-concentratord automatically

ChirpStack Gateway OS
1

Hi there,

I am running Gateway OS on the PI with RAK831, I got a problem in which the chirpstack-concentratord does not generate any uplink packet until I manually restart it.

My question is:

  • How can I fix it?
    or
  • Are there any way to configure (either on chirpstack-gateway-bridge or chirpstack-concentratord) to detect this situation and restart the service automatically.

Thanks,

2

I think it would be more important to find out why it doesn’t receive any data instead of automatically restarting it. Could it be that you have configured a gateway-profile with invalid config in ChirpStack? This would cause ChirpStack to push invalid config to the Concentratord, e.g. making it listen to the wrong frequencies. Restarting the Concentratord will make it re-read the config from disk (until new config is pushed by ChirpStack).

1 Like
3

Thank @brocaar,

The following is the log from chirpstack-concentratord:

Nov 29 00:06:51 loragw user.info chirpstack-concentratord-sx1301[488]: Publishing stats event, stats_id: c3b5cb3d-27bc-4470-9b3a-477a9778ad66, rx_received: 6, rx_received_ok: 5, tx_received: 0, tx_emitted: 0
Nov 29 00:07:03 loragw user.info chirpstack-concentratord-sx1301[488]: Frame received, uplink_id: 6695a70f-234b-41ee-bb60-569724387228, count_us: 2870581307, freq: 923400000, bw: 125000, mod: LoRa, dr: SF7
Nov 29 00:07:07 loragw user.info chirpstack-concentratord-sx1301[488]: Frame received, uplink_id: 4bbdecb3-b326-4659-a184-a1f057a1aed0, count_us: 2874610051, freq: 923400000, bw: 125000, mod: LoRa, dr: SF7
Nov 29 00:07:09 loragw user.info chirpstack-concentratord-sx1301[488]: Frame received, uplink_id: b71f0f0c-64b5-443b-b203-85f0c6bdfc4d, count_us: 2875931731, freq: 923400000, bw: 125000, mod: LoRa, dr: SF7
Nov 29 00:07:21 loragw user.info chirpstack-concentratord-sx1301[488]: Publishing stats event, stats_id: bca0f28b-4336-4aa4-b27e-4fab62f55953, rx_received: 3, rx_received_ok: 3, tx_received: 0, tx_emitted: 0
Nov 29 00:07:51 loragw user.info chirpstack-concentratord-sx1301[488]: Publishing stats event, stats_id: 0362e2b9-b9ab-4a47-845e-b183477de3f2, rx_received: 0, rx_received_ok: 0, tx_received: 0, tx_emitted: 0
Nov 29 00:08:21 loragw user.info chirpstack-concentratord-sx1301[488]: Publishing stats event, stats_id: 02e6ff12-12af-4602-a4e1-0e708271dbbf, rx_received: 0, rx_received_ok: 0, tx_received: 0, tx_emitted: 0

I could see that the RAK831 RX LED has been turned off after chirpstack-concentratord did not receive any uplink, is this a problem with a crappy Lora card?

I did not make any special configuration on Chirpstack Network for this test, in this case is only the region name (which is AS923). Every time I restart the concentratord, it can work in about 6 hours