I wrote about this in late 2023 but no action was taken.
I got more insight into the problem.
Scenario:
- 2 Hiveon T17 running, A and B. Same version. Latest version circa mid 2023. Same pools.
- 1st pool is valid. I routed the connection to the 1st pool through ISP B.
- 2nd and 3rd pools are valid, routed through ISP A.
- I waited 1 day, everything OK.
- RB750 router.
- I disconnected ethernet cable to ISP B for 30 seconds.
- One Hiveon signaled “all pools dead” and restarted after 45 seconds.
– It did not recover when the cable was plugged back. - The other Hiveon did not signal anything. Continued running.
- Other firmwares/machines continued running.
Is dev fee being paid for nothing?
There are more ill bugs that become evident when there are network difficulties.
For example, pool IPs and fee IPs are not checked before warming up the chips.
It warms up the chips and then checks IPs. It keeps rebooting if IPs are not reachable.
But this is the main one. Please fix this.
*** At some point in the connection state machine, if the connection to active pool gets interrupted with a certain error code (like unreachable), it thinks all the pools are dead!
Regards