This is v1.1.1 of Hyperdrive. This fixes some issues reported by the community, and adds a few important features related to upgrading.

This is a high priority update for anyone that switched from Holesky to Mainnet on the same machine; in general, all users should plan to update when they are able to do so.

Known Issues

• Lodestar v1.22.0 does not currently support x64 systems without the avx instruction set. This will be fixed in a later client release.

• Reth still has a configuration issue that prevents it from working with the StakeWise v3 Operator service. If you're a StakeWise operator, be advised that you will have to use a different client for now.

  • We believe we have a fix in place for this and will release it in a future version.

Hyperdrive Core

Clients

• Updated Besu to v24.10.0.

• Updated Nethermind to v1.29.1.

• Updated Reth to v1.1.0.

• Updated Nimbus to v24.10.0.

• Updated Prysm to v5.1.2.

• Updated Teku to v24.10.3.

• Updated Prometheus to v2.55.0.

Service

• Installing Hyperdrive updates via apt update now prompts you to automatically restart the service so the daemons will be up to date.

• Automatic TXs will now ignore the gas threshold setting if a manual max fee is specified, as the threshold setting is only useful for dynamic network-based fees.

• Automatic TXs will now lower the max priority fee down to the max fee if the priority fee is greater than the max fee.

Config

• Switching networks in the service config network dropdown now properly loads the new network settings before starting the containers again.

  • Thanks to @jshufro for reporting this issue!

• The default max fee has been reduced to 20, and the default max priority fee has been reduced to 0.1. Note this only affects new users that are configuring Hyperdrive for the first time.

• The config validation routine will now work properly if you don't have any of the default MEV-Boost relays enabled but do have your own custom relays.

• Changing the contents of the MEV-Boost custom relay option will now properly flag the mev_boost container as needing to be restarted.

CLI

• Updating manually via hyperdrive service install will now automatically restart the service unless the --no-restart flag is provided.

• New global flags:

  • --itsf can be used to ignore TX simulation failues and sign / submit TXs anyway, even if they will revert.
  • --fgl can force a specific gas limit on all transactions created by whatever command is being run.
  • Do NOT use these unless you know what you're doing and have a good reason.

• The max fee selection prompt now properly shows the low and high transaction cost range, instead of just showing the high cost twice.

• Running Constellation or StakeWise commands without the respective module enabled will now print a proper warning message instead of a connection error.

• Manual TXs will now lower the max priority fee down to the max fee if the priority fee is greater than the max fee, and prompt the user with a notice before confirming the fee choices.

Modules

Constellation

• Fixed a bug that broke Constellation's automatic signed-exit upload task if an exit message was manually uploaded first via cs minipool upload-signed-exit.

• Fixed a bug that caused Rocket Pool contract queries to hang on single-CPU systems.

• New flags on cs minipool create:

  • --sbc skips checking if the node balance is sufficient to cover the temporary staking lockup.
  • --slc skips checking if Constellation has enough ETH and RPL in its vaults to fund a new minipool.
  • Only use these if you intend to sign (but not submit) a TX for manual bundling, or submit at a later date when the on-chain conditions become correct. Otherwise the transaction may revert when submitted and cost you gas fees!

• Removed delegate info from cs minipool status since it is not relevant for Constellation node operators.