Hello ChirpStack team,
We are observing an abnormal backhaul data usage issue with one specific gateway after migrating from ChirpStack v3 to v4.
Scenario:
Network: ChirpStack v4 (Docker-based deployment)
Gateway transport: MQTT over cellular
All gateways are configured similarly and have the same number/type of sensors
Sensors transmit standard uplinks (very low LoRaWAN payload size)
Issue:
One gateway (647fdafffe01358e) is showing extremely high cellular usage:
~600 MB downlink
~300 MB uplink
Whereas other gateways in the same setup show normal usage:
~30 MB down / ~15 MB up
~10 MB down / ~10 MB up
Observations:
LoRaWAN uplink/downlink frame sizes are very small (verified from Frames tab)
No unusual sensor traffic compared to other gateways
Gateway is operational and forwarding data correctly
Issue is isolated to a single gateway
Suspicions / questions:
Can repeated MQTT/TLS reconnects or unstable backhaul cause excessive downlink traffic?
Does ChirpStack re-send configuration, stats acknowledgements, or state sync repeatedly on reconnect?
Are there known cases where region mismatch or ADR instability leads to excessive gateway traffic?
Are there recommended ways within ChirpStack to:
Detect frequent gateway reconnects?
Reduce control / stats traffic?
Tune gateway stats or keepalive behavior?
We would appreciate any guidance on:
Known causes for such asymmetric high backhaul usage
Logs or metrics we should specifically inspect on the ChirpStack side
Best practices to avoid this scenario
Thank you in advance for your help.
Have you looked at the gateway logs to see if there is anything abnormal? Given that both the uplink and downlink numbers are higher than the other gateways, I assume this is not caused by neighboring network(s) / devices that your gateway picks up but are not part of your network.
Based on the GatewayID you gave this is a Tektelic Kona Mega, are all your other gateways the same model?
Just for comparison I have a Konn Mega that is processing ~45000 uplinks/13000 downlinks per day, and uses around 100mb per day of total data. How does your gateway compare in terms of that?
If the traffic is not from chirpstack packets could someone be SSHed into the device causing the extra data? If you SSH into the gateway and run the command 'netstat -tnpa' it will show the current active connections. compare the results to a lower data use gateway.
Hey @brocaar , I have looked into the gateway logs. But all were clear and similar to the other gateways after configuration. I don't understand why we are still getting so high uplink and downlink. Is it normal to get such high uplink/downlink ? Client said there is a similar gateway and sensor set in different area but its uplink/downlink are like in the range of 10-20 MB. So I'm concerned, can you please help me with this.