Statement distribution likely does not connect properly to the validator peer set.

[eskimor]

See discussion here

[eskimor]

Ok, it will take me a bit longer to look into this. If anyone wants to pick this up sooner - be my guest :-) It is not super critical as connections seem to work since #2372 .

[rphmeier]

Statement distribution doesn't need to connect to the validator set as it's meant to be a gossip protocol where all validators see all statements. The statements themselves are light in the case of code upgrades. although we should optimize this later on.

The reason all validators need to see all statements is because the block author needs to include backed candidates from all parachains, not just its own group. And in BABE or SASSAFRAS (future), the block author isn't known to other nodes.

[rphmeier]

What this "bughunting" issue is really about is ensuring that statements are propagating correctly among all nodes.

[eskimor]

In particular it should be checked that connections on the validation peer set are actually opened before messages are being sent to a particular peer.

[eskimor]

Closing, as obsoleted by #2536 .