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
As of this PR we have some warning when authority lookup fails for too many peers over a longer period of time. Being connected to validators is essential for a validator, so we would like to know if there are problems.
The above mentioned PR does not go far enough unfortunately, as it only tracks whether an authority could be looked up, but does not track actual connections. For this gossip-support should keep track of PeerConnected and PeerDisconnected events as well and report warnings if there are not enough connections.
Another issue with that PR is, that the monitoring is not per session but for a couple of sessions combined. So the 2/3 threshold is not that meaningful here. Actual connection tracking should keep track of connectivity per session we are interested in and issue warnings whenever connectivity on one of those sessions goes below a threshold.
The text was updated successfully, but these errors were encountered:
* adds transfer token test
* assert agent id is the source
* complete transfer token test
---------
Co-authored-by: claravanstaden <Cats 4 life!>
Co-authored-by: Alistair Singh <alistair.singh7@gmail.com>
helin6
pushed a commit
to boolnetwork/polkadot-sdk
that referenced
this issue
Feb 5, 2024
As of this PR we have some warning when authority lookup fails for too many peers over a longer period of time. Being connected to validators is essential for a validator, so we would like to know if there are problems.
The above mentioned PR does not go far enough unfortunately, as it only tracks whether an authority could be looked up, but does not track actual connections. For this
gossip-support
should keep track ofPeerConnected
andPeerDisconnected
events as well and report warnings if there are not enough connections.Another issue with that PR is, that the monitoring is not per session but for a couple of sessions combined. So the 2/3 threshold is not that meaningful here. Actual connection tracking should keep track of connectivity per session we are interested in and issue warnings whenever connectivity on one of those sessions goes below a threshold.
The text was updated successfully, but these errors were encountered: