[network-health] Metrics page incorrectly showing downtime

Hi - I am having an issue where the Tor Metrics page for my [bridge](https://metrics.torproject.org/rs.html#details/FE31AD644BA16316A06B905F8403F...) will always eventually be showing between 1-2hr downtimes, despite the server showing no notable logs, being reachable, and displaying running circuit/data activity using Nyx.

I have tried rebooting the machine and the service, which at best fixes it for a week before the displayed downtimes begin.

I am pretty confident in the operation of the bridge and am worried that traffic may be steered away due to incorrect descriptors going around.

Best, Hakeem