I received my apollo-2 full node systems about a month ago,
and it's been quite an educational journey! I joined the
ocean pool and received my first payout with lightning.
However a few days ago the bitcoin nodes on both my apollo-2
systems (configured as TOR mode) lost connection to other nodes,
so they're stuck syncing blocks (~644000/844366).
Tor logs in /var/log/syslog are filled with messages like this :
May 22 04:49:35 futurebit-apollo-2 Tor[202872]: No circuits are opened. Relaxed timeout for circuit 123467 (a Hidden service client: Fetching HS descriptor 4-hop circuit in state waiting to see how other guards perform with channel state open) to 60000ms. However, it appears the circuit has timed out anyway. [1348 similar message(s) suppressed in last 3900 seconds]
May 22 09:42:24 futurebit-apollo-2 Tor[202872]: We'd like to launch a circuit to handle a connection, but we already have 32 general-purpose client circuits pending. Waiting until some finish. [4454 similar message(s) suppressed in last 600 seconds]
May 22 09:47:18 futurebit-apollo-2 Tor[202872]: Failed to find node for hop #1 of our path. Discarding this circuit.
The internet connection is fine, and the miner continues to contribute
to the ocean pool.
I am able to use tor from other computers on my LAN.
Rebooting the apollo-2 does not help.
Tor version 0.4.6.10 is a couple years old.
How are updates managed?
Do you have any hints about this?
Thanks!
Tor is notorious for sync issues. I would temporarily turn tor off and let the node sync then turn it back on.
Update issues have been resolved, so as long as you dont do a dist upgrade you should be fine with apt update/upgrade going forward (kernel is now frozen).