Sally MacFarlane
Sally MacFarlane
also noticed on 6609-00 and 6609-02, where I forcibly added the 2 bootnodes as peers, those nodes haven't added any other peers via discovery (over last 9hours) - whereas other...
with no further intervention, 6609-90 is now up to 25 peers 
agree @Beanow - I'm actually working on a related fix right now that will help a lot to get rid of UNKOWN disconnects
@Beanow see #6968 which gets rid of majority of UNKNOWN disconnects. can then do a follow on refactor that changes the method signature as you suggest
useful command to collect data on disconnect reasons by client name `cat /var/log/besu/besu.log | grep -v MISMATCHED_NET | grep "bound - 0x" | awk '{print $16,",",$19,$22}' | sort` example output...
maybe we want to throttle this, depending on how spammy it is?
going to test this on some nodes and see how spammy it is
combined this with changes from #6831
> going to test this on some nodes and see how spammy it is too spammy at info level. changed to debug.
this one is still flaky NewPendingTransactionAcceptanceTest > subscriptionToMinerNodeMustReceiveEveryPublishEvent() FAILED eg https://github.com/hyperledger/besu/actions/runs/8769973760/job/24066042552 at app//org.hyperledger.besu.tests.acceptance.pubsub.NewPendingTransactionAcceptanceTest.subscriptionToMinerNodeMustReceiveEveryPublishEvent(NewPendingTransactionAcceptanceTest.java:132)