You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 15, 2023. It is now read-only.
In case of a finished dispute, we no longer want to communicate with validators which have been slashed significantly. We should make sure to remove those validators from the reserved set and also ban any connection attempts via free incoming slots.
For request/response we should probably stop using IfDisconnected::TryConnect, so disconnecting validators from peersets is enough to stop any communication with them.
The text was updated successfully, but these errors were encountered:
There is a notion of 'disabled' validators in the FRAME staking / session pallets. We should extend the implmentation to cover the remainder of the current era and communicate 'disabled' validators to the network bridge or other networking subsystems so they can ignore messages.
Ideally, there'd be a round-robin mechanism so if more than f validators are disabled that we start allowing communication from the remaining 2f+e which is necessary to maintain consensus. Otherwise, we risk a liveness error.
In case of a finished dispute, we no longer want to communicate with validators which have been slashed significantly. We should make sure to remove those validators from the reserved set and also ban any connection attempts via free incoming slots.
For request/response we should probably stop using
IfDisconnected::TryConnect
, so disconnecting validators from peersets is enough to stop any communication with them.The text was updated successfully, but these errors were encountered: