The result json has been changed in this way:
- Renamed two fields:
- bestEpochNumber -> bestBlockEpochNumber
- bestSlotNumber -> bestBlockSlotNumber
- Added these two new fields:
-
currentEpochNumber
-
currentSlotNumber
(their value is calculated based on the current timestamp, while the previous two were and are calculated based on the best block available in the history)
-
New configuration option onlyConnectToKnownPeers
was added to network
section of configuration file. When set to true
it will force node to only try connecting to known peers. This option is disabled by default.
sparkz {
network {
onlyConnectToKnownPeers = false
}
}
This feature adds a dedicated connection pool reserved for connecting to forger nodes. The size of the dedicated connection pool is configured by maxForgerConnections
option in network
section of configuration file. The default value is 20. This limit works in addition to the one defined in the property network.maxOutgoingConnections.
Configuration option isForgerNode
in the network
section has to be set to true
to indicate that the node is a forger and other nodes should prioritize connecting with it. Default value is false
.
Nodes will prioritize connecting to and broadcasting blocks to forger nodes.
sparkz {
network {
maxForgerConnections = 20
isForgerNode = false
}
}
The intent of this feature is to minimize block propagation time between forgers, by increasing the likelihood of 1-hop connectivity between forgers.
- delete
peers
directory from thesparkz.dataDir
location - update configuration file with new options, if any new parameter is different from the default:
sparkz.network.onlyConnectToKnownPeers
sparkz.network.maxForgerConnections
sparkz.network.isForgerNode
- Synchronization race condition fix
- Block synchronization performance improvements:
- new block lookup strategy based on the parent block id
- preserve ordering of block during synchronization
- history comparison logic rework
- other performance improvements
Full Changelog file here