I noticed while testing that I could start up multiple RedisBungee instances of the same id with no explicitly thrown errors. I figured there must already be a system in place to handle that, and after looking at it I found the source of the issue.
Even though the repeating heartbeat task was setting the timestamp in seconds, both the initial heartbeat set on plugin start, and the processing of the impostor RedisBungee instance were using milliseconds.
This PR fixes everything heartbeat-related to use the timestamp in seconds.
I noticed while testing that I could start up multiple RedisBungee instances of the same id with no explicitly thrown errors. I figured there must already be a system in place to handle that, and after looking at it I found the source of the issue.
Even though the repeating heartbeat task was setting the timestamp in seconds, both the initial heartbeat set on plugin start, and the processing of the impostor RedisBungee instance were using milliseconds.
This PR fixes everything heartbeat-related to use the timestamp in seconds.