Guys, sorry if anyone took offence to initial post…guess my humour fell flat, in response to Nick. As noted there is a FUP in play for anyone utilising the TTN infrastructure, though we all stretch a point now and again … both backend and the GW/network deployments (am sensitive on last point given have donated literally dozens of GW’s to the communities use.)
I quite enjoy the tests and antics of the HAB community within the context of LoRa & LoRaWAN, and followed and even encouraged some use from early days by the likes of Stuart and Dave Ack…et al… as I’m sure @LoRaTracker would confirm.
From FUP perspective yes it’s 30s uplinks per day, what isn’t obvious and I guess no one wants to get into specifying minutiae and details, is this is predicated on typical sensor use cases we see across the network where a sensor may well be seen by one or two or three local GW’s, poss a hand full or low teens in high density and some urban environments or where sensor is placed high with good view of many GW’s in surroundings. I have a few sensors myself which regularly see 10, 12, even 15 gw’s.
HAB users really should consider that in some cases (out of and control, give vagaries of the weather and wind, and even ducting effects) balloon message could be seen by dozens and even hundreds of GW’s each time… remember the backend has to deal with de-duping this peak load. Also none of us mind More extreme hits on a gw occasionally…what’s 60, 90, 180, 400ms between friends on a gw?! In more critical areas logic dictates deploy redundant GW’s in case of blocking on one. That way the users can have greater assurance of message getting through somehow.
Problem I realised during one HAB run a while back is that with height comes ability to blanket hit all the GWs in an area … and extensive one at that. IIRC there was one run pre-pandemic where gw’s across S.E England, N.France, Belgium, The Netherlands, Lux, even into Germany and again IIRC part of Denmark, though perhaps not all at same time…maybe it was across part of the track, but either way >>500gws were being pinged each time. Inc a half dozen of my own, loosing their inherent redundancy… just sayin, just food for thought!
FUP may allow for aro 3 min intervals in this case @ SF7, vs 30s, but as ever when considering LoRaWAN, and TTN, question should always be asked by users, “just ‘cause I can does that mean I should”!
In this case how would the local, SWIN and more importantly the TeviNET infrastructure providers and community of users down in the S.West feel if they knew there was even a, partial, DoS? We don’t want such users discouraged from deploying into & expanding TTN and going private instead. Again just saying
It’s a personal view, again as a GW provider, but I would respectfully ask that there is some backoff even again FUP, given the more extreme use cases represented by HAB…