BigEnigma
commented
2 years ago Ignore this was not the case, mislead by naming in the code. So there is some kind of 10 minute periodic check to update status which also seems to be limited to the numbers handled each cycle....which leads to some additional delay in the status being updated. This is compounding the latency in heartbeats and the subsequent reactivation period such that users might get notified of going offline and then for whats slowly becoming a larger variable period of time it might then become activated, the end user left wondering should they reboot or not
THIS DEPENDS ON THE SEMANTICS OF ONLINE, ACTIVE ETC
The online status reflected in gateway_status, and therefore the api and the explorer website is being derived from the true activities of a hotspot, and does not monitor the last_poc_challenge value so its not actually changing the status reported when a hotspot is merely reactivated via the validator heartbeat which simply bumps that value.
https://github.com/helium/blockchain-etl/blob/master/src/be_db_gateway_status.erl#L207