Appearance

v0.0.1-rc9 ​

This release supersedes v0.0.1-rc8.

Highlights ​

  • Improved fork digest discovery: Enhanced fork digest detection with fallback mechanisms and better logging. The gateway now checks its own subscribed topics if peer discovery fails, improving reliability during network transitions.
  • Consolidated topic management: Topic discovery and refresh operations are now handled centrally, reducing missed or duplicated topic subscriptions and improving stability of topic membership.
  • Batched unsubscribe handling: Improved unsubscribe operations to avoid noisy immediate unsubscriptions and provide more stable topic membership during fork transitions.
  • Removed automatic restart routine: The gateway no longer performs periodic restart signaling for the CL node, simplifying operations and reducing unnecessary interactions.
  • Improved dashboard: Enhanced Grafana dashboard with better clarity, corrected negative values, and improved data visualization for latency metrics.

Configuration update ​

Gateway Cluster ID Change: RC9 requires updating gateway_cluster_id from optimum_hoodi_v0_1 to optimum_hoodi_v0_2:

yaml
gateway_cluster_id: optimum_hoodi_v0_2  # Updated from optimum_hoodi_v0_1

Important: This change is required for proper network connectivity. Gateways using the old cluster ID will fail to connect to the network.

Bug Fixes ​

  • Fork digest discovery reliability: Added fallback detection using gateway's own subscribed topics when peer discovery fails, ensuring fork digest changes are detected even when peer messages are delayed or missing.
  • Topic subscription stability: Consolidated topic discovery and refresh operations to prevent missed subscriptions and duplicate topic operations during fork transitions.
  • Unsubscribe handling: Batched unsubscribe operations to avoid immediate unsubscriptions that could cause topic membership instability.
  • Dashboard improvements: Fixed negative values in latency metrics and improved data visualization clarity.

Known Issues ​

  • Lighthouse connection stability: Some Lighthouse clients may experience connection issues. This will be addressed in RC10.
  • Prysm client configuration: It is recommended to run 2 Prysm clients for the time being for proper configuration.

Migration from RC8 ​

Required configuration change:

Update your gateway_cluster_id in app_conf.yml:

yaml
gateway_cluster_id: optimum_hoodi_v0_2  # Changed from optimum_hoodi_v0_1

Then restart your gateway:

bash
docker pull getoptimum/gateway:v0.0.1-rc9
docker restart optimum-gateway

No other configuration changes are required. All other RC8 features continue to work unchanged.