Closed jp-bennett closed 1 month ago
@thebentern this will result in no time source availability in a mesh where no GPS nodes are present. I would suggest to still allow NTP otherwise this is a pretty drastic change.
Yes, I think NTP quality is probably a good compromise of allowing time source injections while avoiding poor quality times. Care to make a PR?
@thebentern this will result in no time source availability in a mesh where no GPS nodes are present. I would suggest to still allow NTP otherwise this is a pretty drastic change.
Here you go: https://github.com/meshtastic/firmware/pull/4068
I'm seeing a situation where a "rogue" node thinks it has a valid timestamp, but doesn't. It's using fixed position, so the wrong timestamp is overwriting a time provided by another node that actually has a GPS.