What happens if a device exceeds the fair use policy ?
I cannot find in documentation what is the “fine” if a device exceeds the rules of fair policy.
Will it be blocked for a certain time ?
thanks in advance,
anton
The “fine” is your account is suspended. One glance at the graphs by the TTI SysOps, one click to suspend. Sure, you can setup another. But that will get suspended too, until you get bored.
For what I hope are blindly obvious reasons, this isn’t documented or discussed in detail otherwise people spend time trying to workaround the FUP or just debate the details.
The bottom line is that this is a service funded by TTI for the use of a global community who’s only contribution to the running costs is hosting gateways. If we stress the system too much and incur additional costs for TTI they may have to put some realtime rate limiting or
Someone even thinking about getting close to the limits of the FUP is rather worrying.
If you think you need to use more than your fair share, then perhaps pay for an instance of your own where you are limited by local legal constraints.
A device can use up its allocation in fairly short order - so a lot of uplinks over a space of a couple of hours - as it’s a 24 hour policy. You can’t “share” time or downlinks with other devices.
If you are developing a device there is no need to soak test anything - you can test the radio with a push-to-send button, you can test your backend by using the console to “send” an uplink and you can test your code by printing the payload as hex and copying & pasting that in to the console to see how it decodes / hits your backend.
If you are crazy enough to be writing your own LoRaWAN stack, then you can get a free Discovery instance from TTI to test against. But the reality is with test driven development even that’s not strictly needed.
Most people get stuck on the number of downlinks - if you need more than 1 a fortnight you probably want to discuss your use case on the forum here so we can help find a better way.
1 Like
thank you very much for your extended answer to my question.
the reason: i was testing a new aplication that does an upload every 60 seconds with a payload of about 12 bytes.
this test was running for about a day without any problems but suddenly the mqtt sesion was interrupted(and restarted) and after this point i see the application data in the console but my mqtt sessions keeps waiting.
So i was afraid that i was maybe violating the FUP.
I have restarted my application with with much lower rate ( 1 upload every 2 hours) but still the same: after joining the data message is accepted (Forward uplink message) but my receiving mqtt session does not receive anything.
You were …
Going from once per minute to once every 120 minutes makes for some unusual tests - what were you trying to do?
Are uplinks showing in the console? What MQTT client or code base are you using? How many instances or copies of the client are running - there’s no FUP as such on MQTT but it’s one of the larger loads on the infrastructure and peeps running many clients do find some variability on TTN - if you have several running, then web hooks are preferable or a single MQTT connection to your own database where you can then run your own broker.
to answer your questions:
- uplink messages are visible in the console
- mqtt client: mosquitto on debian 12 LXC
- 1 instance which replicates messages to the local mqtt environment for further processing
anyway, after 24 hours of silence from mqtt-ttn everything came alive again.
So in my experience it looks that if I exceed the FUP “the punishment” is 24 hours silence from TTN for that application.
(another application which runs already several years was not interrupted)
on your advice i installed a webhook and this is also working perfectly.
so i will switch to the webhook solution.
many thanks again for your attention on this matter.
This topic was automatically closed 24 hours after the last reply. New replies are no longer allowed.