Adrian-Stefan Mares

We also have `ttn-lw-cli use` in order to generate the configuration file - is that slower than clicking the button in the Console and moving the file to the right...

> The `class_c_always_enabled` flag is simply added to the end device itself. It serves as a default for `class_c_enabled` in a new MAC state. Again, only for 1.0.x devices. In...

> A previous work around to specify the extra channels as normal uplink-channels with only a single data rate is no longer supported due to #6182 Could you please give...

So you guys were ahead of your times, sort of, and this issue (#6526) is just a symptom of your real problem (#6527). For pretty much the whole existence of...

> @adriansmares indeed we are working outside the fixed band plan intended limits. We don't specifically need the NS to send down `NewChannelReq` however if an uplink arrives to the...

In that case the two proposed changes in https://github.com/TheThingsNetwork/lorawan-stack/issues/6526#issuecomment-1715115005 should suffice - please let me know if that is not the case.

We should also take a look at the scheduling event error propagation. Right now the cause is determined by the error code of the returned errors, and we probably should...

The problem here is that the redirect URI is not provided during the token exchange. The OAuth client which is affected here has multiple redirect URIs, and the first redirect...

> 1. nor is `rx1_delay` validated on maximum value (https://github.com/TheThingsNetwork/lorawan-stack/blob/v3.25/api/lorawan.proto#L508-L509). This means that an RX1 delay of 1 year may be possible. I'm not sure, but I don't see why...

> There's nothing that prevents you from scheduling class C downlink in another sub-band. Unlikely, but correct. > I would start by logging long delays in GS. Anything beyond the...